- Email: [email protected]
Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology
Journal of Clinical Neuroscience xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology Graeme A. Brazenor a, Gregory M. Malham a,⇑, Zita E. Ballok b a b
Neuroscience Clinical Institute, Epworth Hospital, Melbourne, VIC, Australia Nuclear Medicine Department, Primary Healthcare Imaging, Epworth Hospital, Melbourne, VIC, Australia
a r t i c l e
i n f o
Article history: Received 13 August 2013 Accepted 16 November 2013 Available online xxxx Keywords: Bone scan CT scan Diagnosis MR Pathology Spine Treatment
a b s t r a c t Image fusion software enables technetium99m-methylene diphosphonate (Tc99m-MDP) bone scan images to be co-registered with CT scan or MRI, allowing greater anatomical discrimination. We examined the role of bone scan images co-registered with CT scan or MRI in the investigation of patients presenting with axial spinal pain and/or limb pain. One hundred and thirty-nine consecutive patients were examined, and thereafter investigated with CT scan, MRI, and/or dynamic plain films. At this point diagnosis (pathology type and anatomical site) and treatment intention were declared. The co-registered Tc99mMDP bone scan images were then studied, after which diagnosis (pathology type and anatomical site) and treatment intention were re-declared. This data were then analysed to determine whether the addition of co-registered bone scan images resulted in any change in diagnosis or treatment intention. The most significant change in diagnosis was pathology type (10%). Anatomical site changed markedly without overlap of the pre and post-isotope fields in 5%, and with overlap in 10%. Treatment intention had a major change in 3.6% and minor change in 8.6%. In the two groups where there was (i) no obvious pathology after full pre-isotope investigation, or (ii) a spinal fusion under suspicion, addition of the bone scan information led to a major change in the pathology and/or anatomical localisation in 18% and 19%, respectively. The addition of co-registered Tc99m-MDP bone scan images offers significant diagnostic assistance, particularly in the difficult diagnostic groups where a failed spinal fusion may be the suspected pain generator, or when no pain generator can otherwise be found. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Bone scans are a non-invasive diagnostic modality that use radiolabeled bisphosphonates to identify areas of abnormal osteogenesis. The most widely used radioisotope is technetium99mlabeled methylene diphosphonate (Tc99m-MDP) [1]. The isotope binds to hydroxyapatite at sites of active osteoblast formation. Areas of focal isotope uptake are detected by a gamma camera, identifying pathologies such as degenerative change, fracture, infection, and tumour [2]. Conventional bone scans use planar imaging or tomographic imaging (single photon emission computed tomography [SPECT]). However, these images have low specificity to skeletal anatomy [3]. Recently, image fusion software has been developed to enable images from isotope bone scans to be co-registered with CT scan or MRI [4–6]; this has led to greater anatomic resolution in the co-registered images. ⇑ Corresponding author. Address: Suite 2, Level 1, 517 St. Kilda Road, Melbourne, VIC 3004, Australia. Tel.: +61 3 9866 6650; fax: +61 3 9866 6681. E-mail address: [email protected] (G.M. Malham).
This study aimed to investigate the role of isotope bone scan images co-registered with CT scan or MRI in the investigation of patients presenting to a spine surgeon with axial spinal pain, limb pain, or both. 2. Materials and methods 2.1. Study population This was a prospective, non-randomised study of 139 consecutive patients presenting to a spine surgeon with axial spinal pain, limb pain or both. Patients were interviewed, examined and then investigated with CT scan, MRI and dynamic radiographs. A preisotope scan diagnosis was made with respect to pathology type and anatomic site, and treatment intention was recorded. Each patient’s Tc99m-MDP isotope bone scan images were coregistered with CT scan (or MRI in 19 patients where CT scans were not available) and analysed. A post-isotope scan diagnosis was then made with respect to pathology type and anatomic site, and treatment intention restated.
http://dx.doi.org/10.1016/j.jocn.2013.11.034 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
2
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
The pre and post-isotope scan diagnoses were compared to determine whether addition of the co-registered isotope bone scan images had resulted in any change in diagnosis with respect to pathology type or anatomical localisation, or in the treatment intention. 2.2. Imaging protocol Triple phase bone scans were performed using a dual head gamma camera (e.cam, Siemens, Erlangen, Germany). Following the bolus intravenous injection of 800 megabecquerels (MBq) of Tc99m-MDP, 120 perfusion phase images, each of 1 second duration, were obtained immediately followed by blood pool images for 2 minutes using a 256 256 pixel matrix. Delayed static images were obtained 3 hours post-injection using a 256 256 pixel matrix for 4 minutes, followed by SPECT images of the same spinal region. SPECT images were acquired in a 128 128 pixel matrix with 60 projections over 360 degrees for 20 seconds per projection. Axial, coronal and sagittal tomograms with a slice thickness of 4 mm were reconstructed using iterative reconstruction (four iterations and eight subsets). Raw data of multislice CT scan (Sensation, Siemens) or MRI (1.5 Tesla, Sigma Excite, GE Healthcare, Little Chalfont, Buckinghamshire, UK) images of the same spinal region were exported to the workstation (e-soft, Siemens) for co-registration. Special care was taken to position all patients similarly for anatomic and functional imaging, including the use of knee rests and pillows. Advanced image fusion software (Syngo, Siemens) was used for co-registration of the cross-sectional functional and anatomic images. No instance of significant mis-registration was detected. Radioisotope uptake was graded as normal, physiological or abnormally increased in relation to anatomic structures, and, in patients with a history of prior spinal surgery, in relation to the fusion hardware. Images were interpreted by an experienced nuclear medicine physician (Z.B.). 2.3. Statistical analysis Statistical analyses included chi-squared testing between groups. Statistical analysis was carried out using the Statistical Package for the Social Sciences version 19.0 (SPSS, Chicago, IL, USA) with statistical significance set at p < 0.05.
frequency, after addition of Tc99m-MDP bone scan images co-registered with CT scan or MRI, is shown in Table 2. Pathology type changed in 14/139 patients (10%). Anatomic site had a major change (without overlap of the pre-isotope and post-isotope fields) in 7/139 patients (5%) and changed with overlap in 14/139 (10%). The treatment intention had a major change in 5/139 (3.6%), and minor in 12/139 (8.6%). The pathological diagnosis changed in 14 patients (10%) after addition of Tc99m-MDP bone scan images co-registered with CT scan or MRI. In these 14 patients, after initial imaging with CT scan, MRI and dynamic plain films, seven patients had no obvious pathology detected, three patients were suspected of having failed fusions, two patients were provisionally diagnosed with degenerative disease, one patient was thought to have an osteoporotic insufficiency fracture, and one patient was thought to have recurrence of C7 myeloma. The revised pathology diagnoses are listed in Table 3. A major change in anatomic localisation after consideration of the isotope scan images occurred in seven patients (5%) (Table 4). There was no overlap of the pre and post-isotope anatomic sites in three cervical, two lumbar and two sacral causative pathologies. All seven patients who had their anatomic localisation changed without overlap also had their pathology type changed. In contrast, only half of the 14 patients with anatomic overlap changed pathology type. Comparing before and after addition of the isotope scan images, treatment intention underwent minor changes in 12/139 (8.6%) and major changes in 5/139 (3.6%) patients. The five patients who underwent major changes in treatment intention are listed in Table 5.
Table 2 Change in pathological diagnosis, anatomical localisation and treatment intention after addition of isotope bone scan images co-registered with CT scan/MRI Change following addition of Tc99m bone scan images
Number
%
Pathology type
14
10
Anatomical localisation Major change without field overlap Change with field overlap
7 14
5 10
Treatment intention Major change Minor change
5 12
3.6 8.6
Tc99m = technetium99m.
3. Results The study cohort comprised 139 patients with a mean age of 68.1 years (range 19–90 years) and 58% were male. Provisional classifications of the patients’ pathologies before isotope bone scans are listed in Table 1. The two most frequent provisional pathologies were spinal degenerative disease in 75 patients (54%) and no obvious pathological diagnosis (after CT scan, MRI and dynamic radiographs) in 39 (28%). The change in
Table 3 Change in pathological diagnosis in 14 patients following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope scan pathological diagnosis
Post-isotope scan pathological diagnosis
No obvious pathology
Paranasal sinusitis Inflammatory cervical arthropathy Enthesopathy of ischial tendons Failed incorporation cervical strut graft Degenerative sacroiliac joint Degenerative L3/4 facet joints Sacral ala fracture
Suspected failed spinal fusion
Degenerative L2/3 facet joints Medial tibial plateau fracture Degenerative C2/3, C7/T1 facet joints
Spinal degenerative disease
Rheumatoid arthropathy C1/2 facet joints Osteoporotic insufficiency fracture L2 Degenerative disease thoracic spine Degenerative disease C3/4 facet joints
Table 1 Pre-isotope scan pathology classification Pathology type
Number
%
Spinal degenerative disease No obvious pathology (after CT scan, MRI, dynamic radiographs) Suspected failed spinal fusion Osteoporotic insufficiency fracture Inflammatory arthropathy Trauma Malignancy Total
75 39
54 28
16 3 3 2 1 139
12 2 2 1.3 0.7 100
Osteoporotic fracture Multiple myeloma (recurrent)
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx Table 4 Major change in anatomical localisation (without overlap) in seven patients, following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope pathology type
Pre-isotope localisation
Post-isotope localisation
Failed spinal fusion
L4–S1 C4–C7 L4–S1
Degenerative L2/3 facet joints Degenerative C2/3, C7/T1 facet joints Medial tibial plateau fracture
–
Degenerative sacroiliac joint
No obvious pathology
–
Sacral ala fracture
Degenerative disease
C5/6
Multiple myeloma
C7
Rheumatoid arthropathy C1/2 facet joints Degenerative C3/4 facet joints
Table 5 Pre- and post-isotope diagnoses and treatment intentions in the five patients for whom treatment intention underwent major change following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope scan diagnosis and treatment intention
Post-isotope scan diagnosis and treatment intention
Degeneration right hip joint > lumbar spine Hip replacement Failed C4-C7 fusion Revision of cervical strut graft No obvious pathology No treatment Degeneration lumbosacral spine Conservative management Failed fusion L4-S1 Conservative management
Degeneration lumbar spine > hip Conservative spinal therapy Degenerative C2/3, C7/T1 facet joints Conservative cervical therapy Fracture left sacral ala Conservative fracture treatment Degeneration left hip Hip replacement Medial tibial plateau fracture Conservative knee therapy
Table 6 Patients grouped by pre-isotope pathology type, showing the change in pathological diagnosis, anatomical localisation (without overlap), or major change in treatment intention, following addition of isotope bone scan images co-registered with CT scan/ MRI Pre-isotope scan pathology type (% of total)
Major change, n (% of total)
p value
Spinal degenerative disease (54%) No obvious pathology (28%) Suspected failed spinal fusion (12%) Osteoporotic fracture (2%) Inflammatory arthropathy (2%)
4/75 (5.3%) 7/39 (18%) 3/16 (19%) 1/3 (33%) 0/3 (0%)
0.02 ns ns ns ns
3
3.1. Illustrative Patient 1 A 38-year-old Caucasian man with a history of L5/S1 discectomy 14 years previously and intermittent drug abuse, underwent L4/5 and L5/S1 posterior lumbar interbody fusion for intractable back pain due to severe two-level disc and facet joint degeneration. The early post-operative course was complicated by the patient taking an overdose of oral paracetamol. He recovered, with his spinal function improving significantly, and under instruction escalated his daily walking to 10 kilometres per day, taking only non-narcotic analgesia. Four months post-operatively he complained of severe left sciatic pain. CT scan showed a normal operative construct and no new injury. However, isotope bone scan images fused with the CT scan revealed a stress fracture of the left sacral ala (Fig. 1). Conservative treatment of bed rest and no sitting for 3 weeks followed by progressive restoration of his walking program led to a graduated return to work. He resumed full time employment by 12 months after the surgery, at which time he was taking no analgesia. 3.2. Illustrative Patient 2 A 55-year-old Caucasian woman with a history of C5/6 and C6/7 anterior cervical discectomy and fusions 15 years previously with sub-total relief of cervical pain and brachial neuralgia for 10 years, re-presented with axial neck pain and minor left upper limb pain without neurological deficit. CT scan and MRI revealed a mild to moderate cervical kyphosis, with moderately severe degenerative retrolisthesis at C3/4, and mild degenerative anterolisthesis at C4/5 and C7/T1. The patient underwent cervical corpectomy at C5, C6 and C7, with distraction fusion at C4–T1 using titanium rod and buttress prosthesis [7]. Post-operatively she mobilised well and ceased all analgesia after 3 months. The patient re-presented at 2.5 years post-operatively with recurrence of suboccipital pain without neurological deficit. CT scan showed the titanium prosthesis in good position, but there were subtle linear lucencies between the buttress ends and the adjacent vertebral bone suggestive of micromovement from non-union. However, image fusion showed intense radioisotope uptake in the right C2/3 facet joint (Fig. 2). CT scan guided injection of steroid into the facet joint significantly improved the patient’s pain, and further surgery has not been necessary.
ns = not statistically significant.
4. Discussion Analysis of the patient groups in which the addition of isotope scans produced a change in pathology type, major change in anatomic localisation without overlap, and major change in treatment intention is provided in Table 6. Three pre-isotope scan pathology classifications appeared to benefit from the addition of the isotope scan information: (i) where no obvious pathology was detected (after CT scan, MRI and dynamic plain films); (ii) where a failed spinal fusion was under suspicion; and (iii) those suffering from osteoporotic fractures. In 18%, 19% and 33% of the patients in these three groups, respectively, addition of the isotope scan information resulted in major changes to the diagnosis, treatment, or both. Low cohort numbers meant that the only group statistically significantly different from the others was the large degenerative group (n = 75).
In the clinical and radiological assessment of patients presenting with pain in the spine, limbs, or both, the question of limb or limb girdle pathology versus referral of pain from a spinal source frequently arises. In addition there is always the possibility of dual pathology, adding to the diagnostic difficulty. Does the patient with neck and shoulder pain have cervical spondylosis plus shoulder or acromioclavicular degeneration, or is the shoulder pain referred from the spinal pathology? In the patient with chronic low back pain, does the recent sacroiliac pain radiating to groin and knee come from an upper lumbar disc herniation, or from pathology within the hip joint? Such diagnostic dilemmas beset the musculoskeletal physician and spine surgeon as much as the rheumatologist and hip surgeon, and the answers are not always clear from rigorous history taking and skilful examination. Isotope bone scans can be of particular utility in such cases, and in our series changed the pathology diagnosis in 14/139 patients, and changed the anatomical location of
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
4
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
Fig. 1. Patient 1. (Column 1) Coronal isotope bone scan (top) and CT scan (bottom) of Patient 1. Isotope bone scan fused with lumbar CT scan demonstrates a stress fracture of the left sacral ala (Column 2 = axial, Column 3 = sagittal and Column 4 = coronal). (Column 5) Additional axial images.
the diagnosis without overlap of the pre-isotope diagnostic location in seven of those 14 patients. The major problem with isotope bone scanning in the past has always been poor anatomical resolution of the images. Early studies of isotope scanning for bone-related spinal pathology, using static planar images without SPECT or image fusion, showed that bone scintigraphy failed to identify a number of abnormalities such as pseudoarthrosis [8]. On planar scintigraphy increased activity arising from the posterior vertebral elements is superimposed on
activity from the vertebral body, making localisation of the abnormality difficult [9]. The addition of SPECT improves the utility of the isotope bone scan in detecting pseudarthrosis [10], sources of post-surgery spinal pain [11] and spinal osteomyelitis [12]. Image fusion is the process of image superposition using two or more different sets of images. This process provides the functional information in an anatomical context provided by the CT scan. Typically one anatomical image set, (most commonly CT scan, but
Fig. 2. Patient 2. (Column 1) Coronal CT scan of the cervical spine showing the titanium prosthesis in good position, but linear lucencies between the buttress ends and the adjacent vertebral bone were suggestive of micromovement from non-union. Isotope bone scan and CT scan image fusion shows intense radioisotope uptake in the right C2/3 facet joint (Column 2 = axial, Column 3 = sagittal and Column 4 = coronal).
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
occasionally MRI) is combined with functional images from conventional nuclear gamma camera SPECT. The first software algorithms for post-scan image fusion required fiducial markers or bands [13,14], but with subsequent algorithms these were no longer required [15,16]. Such imagefusion programs now allow the functional images to be fused with those of high-resolution diagnostic quality CT and MR scanners. Various authors have reported that the improved anatomical resolution aids diagnosis and staging of gynaecological cancers [4], recurrent prostate cancer [5], lymphoma [15], the sensitivity of scanning for pancreatic tumours [17], the diagnosis of hyperparathyroidism [18], and for thyroid disease [19,20]. Using CT scan/ SPECT image fusion in 37 patients with lumbar pain, McDonald et al. reported that image fusion allows definitive localisation of ‘‘hot’’ facet joints, with localisation superior to that of high resolution SPECT alone [21]. In our study cohort the use of high resolution diagnostic CT scan, MRI, or both, was crucial to assess integrity of fusion hardware, disc prolapse, neural compression, and other pathological conditions not readily demonstrated on isotope bone scan alone. The image fusion program permitted functional images to be superimposed on the diagnostic-quality anatomic images from high definition CT scan or MRI. This enabled highly accurate localisation of the scintigraphic changes to anatomic structures such as facet joints, pars fractures and bony fusion masses (both interbody and intertransverse components). In a small number of patients, more distant pathology became evident with the addition of extended imaging fields that were used according to the symptoms of the individual patient. The alternative to image fusion software is scanning with hybrid SPECT/CT cameras. However, the lower cost hybrid SPECT/ CT cameras produce ‘‘non-diagnostic’’ quality CT images [18,22], but nevertheless remain in common use, particularly in oncological investigation [5,18]. Satisfactory image quality can be obtained using the more expensive ‘‘high-end’’ SPECT/CT cameras, but in patients who may already have undergone conventional CT scanning in the investigation of their spinal problem, the subsequent scan using hybrid camera means that the patient receives unnecessary extra radiation. In our study we found that the same anatomical accuracy can be achieved with the use of fusion programs, creating fusion images using the previously acquired anatomical (CT scan or MRI) data set, thus avoiding further irradiation of the patient. 5. Conclusions Investigation of patients presenting with axial spinal pain, limb pain, or both, to a spine surgeon benefits from the addition of Tc99m-MDP isotope bone scan images co-registered with CT scan or MRI. These co-registered images may lead to change in the pathology classification of the provisional diagnosis in 10% of patients, change in the anatomic localisation of the pathology in 15% of patients and major change in treatment intention in 3.6% of patients. We propose that bone scans co-registered with CT scan or MRI are particularly useful in the difficult diagnostic groups where a previous spinal fusion exists or when conventional imaging is normal.
5
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] Subramanian G, McAfee JG, Blair RJ, et al. Technetium-99m-methylene diphosphonate – a superior agent for skeletal imaging: comparison with other technetium complexes. J Nucl Med 1975;16:744–55. [2] Cook GJ, Gnanasegaran G, Chua S. Miscellaneous indications in bone scintigraphy: metabolic bone diseases and malignant bone tumors. Semin Nucl Med 2010;40:52–61. [3] Even-Sapir E, Metser U, Mishani E, et al. The detection of bone metastases in patients with high-risk prostate cancer: 99mTc-MDP Planar bone scan, singleand multi-field-of-view SPECT, 18F-Fluoride PET, and 18F-Fluoride PET/CT. J Nucl Med 2006;47:287–97. [4] Tsai CC, Tsai CS, Ng KK, et al. The impact of image fusion in resolving discrepant findings between FDG-PET and MRI/CT in patients with gynaecological cancers. Eur J Nucl Med Mol Imaging 2003;30:1674–83. [5] Schettino CJ, Kramer EL, Noz ME, et al. Impact of fusion of indium111capromab pendetide volume data sets with those from MRI or CT in patients with recurrent prostate cancer. AJR Am J Roentgenol 2004;183:519–24. [6] Amthauer H, Denecke T, Rohlfing T, et al. Value of image fusion using single photon emission computed tomography with integrated low dose computed tomography in comparison with a retrospective voxel-based method in neuroendocrine tumours. Eur Radiol 2005;15:1456–62. [7] Brazenor G. Comparison of multisegment anterior cervical fixation using bone strut graft versus a titanium rod and buttress prosthesis: analysis of outcome with long-term follow-up and interview by independent physician. Spine (Phila Pa 1976) 2007;32:63–71. [8] Hannon KM, Wetta WJ. Failure of technetium bone scanning to detect pseudarthroses in spinal fusion for scoliosis. Clin Orthop Relat Res 1977;123:42–4. [9] Johansen JP, Fossgreen J, Hansen HH. Bone scanning in lumbar disc herniation. Acta Orthop Scand 1980;51:617–20. [10] Gates GF, McDonald RJ. Bone SPECT of the back after lumbar surgery. Clin Nucl Med 1999;24:395–403. [11] Lusins JO, Danielski EF, Goldsmith SJ. Bone SPECT in patients with persistent back pain after lumbar spine surgery. J Nucl Med 1989;30:490–6. [12] Love C, Patel M, Lonner BS, et al. Diagnosing spinal osteomyelitis: a comparison of bone and Ga-67 scintigraphy and magnetic resonance imaging. Clin Nucl Med 2000;25:963–77. [13] Scott AM, Macapinlac H, Zhang J, et al. Image registration of SPECT and CT images using an external fiduciary band and three-dimensional surface fitting in metastatic thyroid cancer. J Nucl Med 1995;36:100–3. [14] Mongioj V, Brusa A, Loi G, et al. Accuracy evaluation of fusion of CT, MR, and Spect images using commercially available software packages (SRS PLATO and IFS). Int J Radiat Oncol Biol Phys 1999;43:227–34. [15] Koral KF, Lin S, Fessler JA, et al. Preliminary results from intensity-based CTSPECT fusion in I-131 anti-B1 monoclonal-antibody therapy of lymphoma. Cancer 1997;80:2538–44. [16] Hutton BF, Braun M, Thurfjell L, et al. Image registration: an essential tool for nuclear medicine. Eur J Nucl Med 2002;29:559–77. [17] Lemke AJ, Niehues SM, Hosten N, et al. Retrospective digital image fusion of multidetector CT and 18F-FDG PET: clinical value in pancreatic lesions-a prospective study with 104 patients. J Nucl Med 2004;45:1279–86. [18] Even-Sapir E, Flusser G, Lerman H, et al. SPECT/Multislice Low-Dose CT: A Clinically Relevant Constituent in the Imaging Algorithm of nononcological Patients Referred for Bone Scintigraphy. J Nucl Med 2007;48:319–24. [19] Profanter C, Wetscher GJ, Gabriel M, et al. CT-MIBI image fusion: a new preoperative localization technique for primary, recurrent, and persistent hyperparathyroidism. Surgery 2004;135:157–62. [20] Lemke AJ, Niehues SM, Amthauer H, et al. Clinical use of digital retrospective image fusion of CT, MRI, FDG-PET and SPECT – fields of indications and results. Rofo 2004;176:1811–8. [21] McDonald M, Cooper R, Wang MY. Use of computed tomography-singlephoton emission computed tomography fusion for diagnosing painful facet arthropathy. Technical note. Neurosurg Focus 2007;22:E2. [22] Nömayr A, Römer W, Strobel D, et al. Anatomical accuracy of hybrid SPECT/ spiral CT in the lower spine. Nucl Med Commun 2006;27:521–8.
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Clinical Study
Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology Graeme A. Brazenor a, Gregory M. Malham a,⇑, Zita E. Ballok b a b
Neuroscience Clinical Institute, Epworth Hospital, Melbourne, VIC, Australia Nuclear Medicine Department, Primary Healthcare Imaging, Epworth Hospital, Melbourne, VIC, Australia
a r t i c l e
i n f o
Article history: Received 13 August 2013 Accepted 16 November 2013 Available online xxxx Keywords: Bone scan CT scan Diagnosis MR Pathology Spine Treatment
a b s t r a c t Image fusion software enables technetium99m-methylene diphosphonate (Tc99m-MDP) bone scan images to be co-registered with CT scan or MRI, allowing greater anatomical discrimination. We examined the role of bone scan images co-registered with CT scan or MRI in the investigation of patients presenting with axial spinal pain and/or limb pain. One hundred and thirty-nine consecutive patients were examined, and thereafter investigated with CT scan, MRI, and/or dynamic plain films. At this point diagnosis (pathology type and anatomical site) and treatment intention were declared. The co-registered Tc99mMDP bone scan images were then studied, after which diagnosis (pathology type and anatomical site) and treatment intention were re-declared. This data were then analysed to determine whether the addition of co-registered bone scan images resulted in any change in diagnosis or treatment intention. The most significant change in diagnosis was pathology type (10%). Anatomical site changed markedly without overlap of the pre and post-isotope fields in 5%, and with overlap in 10%. Treatment intention had a major change in 3.6% and minor change in 8.6%. In the two groups where there was (i) no obvious pathology after full pre-isotope investigation, or (ii) a spinal fusion under suspicion, addition of the bone scan information led to a major change in the pathology and/or anatomical localisation in 18% and 19%, respectively. The addition of co-registered Tc99m-MDP bone scan images offers significant diagnostic assistance, particularly in the difficult diagnostic groups where a failed spinal fusion may be the suspected pain generator, or when no pain generator can otherwise be found. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Bone scans are a non-invasive diagnostic modality that use radiolabeled bisphosphonates to identify areas of abnormal osteogenesis. The most widely used radioisotope is technetium99mlabeled methylene diphosphonate (Tc99m-MDP) [1]. The isotope binds to hydroxyapatite at sites of active osteoblast formation. Areas of focal isotope uptake are detected by a gamma camera, identifying pathologies such as degenerative change, fracture, infection, and tumour [2]. Conventional bone scans use planar imaging or tomographic imaging (single photon emission computed tomography [SPECT]). However, these images have low specificity to skeletal anatomy [3]. Recently, image fusion software has been developed to enable images from isotope bone scans to be co-registered with CT scan or MRI [4–6]; this has led to greater anatomic resolution in the co-registered images. ⇑ Corresponding author. Address: Suite 2, Level 1, 517 St. Kilda Road, Melbourne, VIC 3004, Australia. Tel.: +61 3 9866 6650; fax: +61 3 9866 6681. E-mail address: [email protected] (G.M. Malham).
This study aimed to investigate the role of isotope bone scan images co-registered with CT scan or MRI in the investigation of patients presenting to a spine surgeon with axial spinal pain, limb pain, or both. 2. Materials and methods 2.1. Study population This was a prospective, non-randomised study of 139 consecutive patients presenting to a spine surgeon with axial spinal pain, limb pain or both. Patients were interviewed, examined and then investigated with CT scan, MRI and dynamic radiographs. A preisotope scan diagnosis was made with respect to pathology type and anatomic site, and treatment intention was recorded. Each patient’s Tc99m-MDP isotope bone scan images were coregistered with CT scan (or MRI in 19 patients where CT scans were not available) and analysed. A post-isotope scan diagnosis was then made with respect to pathology type and anatomic site, and treatment intention restated.
http://dx.doi.org/10.1016/j.jocn.2013.11.034 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
2
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
The pre and post-isotope scan diagnoses were compared to determine whether addition of the co-registered isotope bone scan images had resulted in any change in diagnosis with respect to pathology type or anatomical localisation, or in the treatment intention. 2.2. Imaging protocol Triple phase bone scans were performed using a dual head gamma camera (e.cam, Siemens, Erlangen, Germany). Following the bolus intravenous injection of 800 megabecquerels (MBq) of Tc99m-MDP, 120 perfusion phase images, each of 1 second duration, were obtained immediately followed by blood pool images for 2 minutes using a 256 256 pixel matrix. Delayed static images were obtained 3 hours post-injection using a 256 256 pixel matrix for 4 minutes, followed by SPECT images of the same spinal region. SPECT images were acquired in a 128 128 pixel matrix with 60 projections over 360 degrees for 20 seconds per projection. Axial, coronal and sagittal tomograms with a slice thickness of 4 mm were reconstructed using iterative reconstruction (four iterations and eight subsets). Raw data of multislice CT scan (Sensation, Siemens) or MRI (1.5 Tesla, Sigma Excite, GE Healthcare, Little Chalfont, Buckinghamshire, UK) images of the same spinal region were exported to the workstation (e-soft, Siemens) for co-registration. Special care was taken to position all patients similarly for anatomic and functional imaging, including the use of knee rests and pillows. Advanced image fusion software (Syngo, Siemens) was used for co-registration of the cross-sectional functional and anatomic images. No instance of significant mis-registration was detected. Radioisotope uptake was graded as normal, physiological or abnormally increased in relation to anatomic structures, and, in patients with a history of prior spinal surgery, in relation to the fusion hardware. Images were interpreted by an experienced nuclear medicine physician (Z.B.). 2.3. Statistical analysis Statistical analyses included chi-squared testing between groups. Statistical analysis was carried out using the Statistical Package for the Social Sciences version 19.0 (SPSS, Chicago, IL, USA) with statistical significance set at p < 0.05.
frequency, after addition of Tc99m-MDP bone scan images co-registered with CT scan or MRI, is shown in Table 2. Pathology type changed in 14/139 patients (10%). Anatomic site had a major change (without overlap of the pre-isotope and post-isotope fields) in 7/139 patients (5%) and changed with overlap in 14/139 (10%). The treatment intention had a major change in 5/139 (3.6%), and minor in 12/139 (8.6%). The pathological diagnosis changed in 14 patients (10%) after addition of Tc99m-MDP bone scan images co-registered with CT scan or MRI. In these 14 patients, after initial imaging with CT scan, MRI and dynamic plain films, seven patients had no obvious pathology detected, three patients were suspected of having failed fusions, two patients were provisionally diagnosed with degenerative disease, one patient was thought to have an osteoporotic insufficiency fracture, and one patient was thought to have recurrence of C7 myeloma. The revised pathology diagnoses are listed in Table 3. A major change in anatomic localisation after consideration of the isotope scan images occurred in seven patients (5%) (Table 4). There was no overlap of the pre and post-isotope anatomic sites in three cervical, two lumbar and two sacral causative pathologies. All seven patients who had their anatomic localisation changed without overlap also had their pathology type changed. In contrast, only half of the 14 patients with anatomic overlap changed pathology type. Comparing before and after addition of the isotope scan images, treatment intention underwent minor changes in 12/139 (8.6%) and major changes in 5/139 (3.6%) patients. The five patients who underwent major changes in treatment intention are listed in Table 5.
Table 2 Change in pathological diagnosis, anatomical localisation and treatment intention after addition of isotope bone scan images co-registered with CT scan/MRI Change following addition of Tc99m bone scan images
Number
%
Pathology type
14
10
Anatomical localisation Major change without field overlap Change with field overlap
7 14
5 10
Treatment intention Major change Minor change
5 12
3.6 8.6
Tc99m = technetium99m.
3. Results The study cohort comprised 139 patients with a mean age of 68.1 years (range 19–90 years) and 58% were male. Provisional classifications of the patients’ pathologies before isotope bone scans are listed in Table 1. The two most frequent provisional pathologies were spinal degenerative disease in 75 patients (54%) and no obvious pathological diagnosis (after CT scan, MRI and dynamic radiographs) in 39 (28%). The change in
Table 3 Change in pathological diagnosis in 14 patients following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope scan pathological diagnosis
Post-isotope scan pathological diagnosis
No obvious pathology
Paranasal sinusitis Inflammatory cervical arthropathy Enthesopathy of ischial tendons Failed incorporation cervical strut graft Degenerative sacroiliac joint Degenerative L3/4 facet joints Sacral ala fracture
Suspected failed spinal fusion
Degenerative L2/3 facet joints Medial tibial plateau fracture Degenerative C2/3, C7/T1 facet joints
Spinal degenerative disease
Rheumatoid arthropathy C1/2 facet joints Osteoporotic insufficiency fracture L2 Degenerative disease thoracic spine Degenerative disease C3/4 facet joints
Table 1 Pre-isotope scan pathology classification Pathology type
Number
%
Spinal degenerative disease No obvious pathology (after CT scan, MRI, dynamic radiographs) Suspected failed spinal fusion Osteoporotic insufficiency fracture Inflammatory arthropathy Trauma Malignancy Total
75 39
54 28
16 3 3 2 1 139
12 2 2 1.3 0.7 100
Osteoporotic fracture Multiple myeloma (recurrent)
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx Table 4 Major change in anatomical localisation (without overlap) in seven patients, following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope pathology type
Pre-isotope localisation
Post-isotope localisation
Failed spinal fusion
L4–S1 C4–C7 L4–S1
Degenerative L2/3 facet joints Degenerative C2/3, C7/T1 facet joints Medial tibial plateau fracture
–
Degenerative sacroiliac joint
No obvious pathology
–
Sacral ala fracture
Degenerative disease
C5/6
Multiple myeloma
C7
Rheumatoid arthropathy C1/2 facet joints Degenerative C3/4 facet joints
Table 5 Pre- and post-isotope diagnoses and treatment intentions in the five patients for whom treatment intention underwent major change following addition of isotope bone scan images co-registered with CT scan/MRI Pre-isotope scan diagnosis and treatment intention
Post-isotope scan diagnosis and treatment intention
Degeneration right hip joint > lumbar spine Hip replacement Failed C4-C7 fusion Revision of cervical strut graft No obvious pathology No treatment Degeneration lumbosacral spine Conservative management Failed fusion L4-S1 Conservative management
Degeneration lumbar spine > hip Conservative spinal therapy Degenerative C2/3, C7/T1 facet joints Conservative cervical therapy Fracture left sacral ala Conservative fracture treatment Degeneration left hip Hip replacement Medial tibial plateau fracture Conservative knee therapy
Table 6 Patients grouped by pre-isotope pathology type, showing the change in pathological diagnosis, anatomical localisation (without overlap), or major change in treatment intention, following addition of isotope bone scan images co-registered with CT scan/ MRI Pre-isotope scan pathology type (% of total)
Major change, n (% of total)
p value
Spinal degenerative disease (54%) No obvious pathology (28%) Suspected failed spinal fusion (12%) Osteoporotic fracture (2%) Inflammatory arthropathy (2%)
4/75 (5.3%) 7/39 (18%) 3/16 (19%) 1/3 (33%) 0/3 (0%)
0.02 ns ns ns ns
3
3.1. Illustrative Patient 1 A 38-year-old Caucasian man with a history of L5/S1 discectomy 14 years previously and intermittent drug abuse, underwent L4/5 and L5/S1 posterior lumbar interbody fusion for intractable back pain due to severe two-level disc and facet joint degeneration. The early post-operative course was complicated by the patient taking an overdose of oral paracetamol. He recovered, with his spinal function improving significantly, and under instruction escalated his daily walking to 10 kilometres per day, taking only non-narcotic analgesia. Four months post-operatively he complained of severe left sciatic pain. CT scan showed a normal operative construct and no new injury. However, isotope bone scan images fused with the CT scan revealed a stress fracture of the left sacral ala (Fig. 1). Conservative treatment of bed rest and no sitting for 3 weeks followed by progressive restoration of his walking program led to a graduated return to work. He resumed full time employment by 12 months after the surgery, at which time he was taking no analgesia. 3.2. Illustrative Patient 2 A 55-year-old Caucasian woman with a history of C5/6 and C6/7 anterior cervical discectomy and fusions 15 years previously with sub-total relief of cervical pain and brachial neuralgia for 10 years, re-presented with axial neck pain and minor left upper limb pain without neurological deficit. CT scan and MRI revealed a mild to moderate cervical kyphosis, with moderately severe degenerative retrolisthesis at C3/4, and mild degenerative anterolisthesis at C4/5 and C7/T1. The patient underwent cervical corpectomy at C5, C6 and C7, with distraction fusion at C4–T1 using titanium rod and buttress prosthesis [7]. Post-operatively she mobilised well and ceased all analgesia after 3 months. The patient re-presented at 2.5 years post-operatively with recurrence of suboccipital pain without neurological deficit. CT scan showed the titanium prosthesis in good position, but there were subtle linear lucencies between the buttress ends and the adjacent vertebral bone suggestive of micromovement from non-union. However, image fusion showed intense radioisotope uptake in the right C2/3 facet joint (Fig. 2). CT scan guided injection of steroid into the facet joint significantly improved the patient’s pain, and further surgery has not been necessary.
ns = not statistically significant.
4. Discussion Analysis of the patient groups in which the addition of isotope scans produced a change in pathology type, major change in anatomic localisation without overlap, and major change in treatment intention is provided in Table 6. Three pre-isotope scan pathology classifications appeared to benefit from the addition of the isotope scan information: (i) where no obvious pathology was detected (after CT scan, MRI and dynamic plain films); (ii) where a failed spinal fusion was under suspicion; and (iii) those suffering from osteoporotic fractures. In 18%, 19% and 33% of the patients in these three groups, respectively, addition of the isotope scan information resulted in major changes to the diagnosis, treatment, or both. Low cohort numbers meant that the only group statistically significantly different from the others was the large degenerative group (n = 75).
In the clinical and radiological assessment of patients presenting with pain in the spine, limbs, or both, the question of limb or limb girdle pathology versus referral of pain from a spinal source frequently arises. In addition there is always the possibility of dual pathology, adding to the diagnostic difficulty. Does the patient with neck and shoulder pain have cervical spondylosis plus shoulder or acromioclavicular degeneration, or is the shoulder pain referred from the spinal pathology? In the patient with chronic low back pain, does the recent sacroiliac pain radiating to groin and knee come from an upper lumbar disc herniation, or from pathology within the hip joint? Such diagnostic dilemmas beset the musculoskeletal physician and spine surgeon as much as the rheumatologist and hip surgeon, and the answers are not always clear from rigorous history taking and skilful examination. Isotope bone scans can be of particular utility in such cases, and in our series changed the pathology diagnosis in 14/139 patients, and changed the anatomical location of
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
4
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
Fig. 1. Patient 1. (Column 1) Coronal isotope bone scan (top) and CT scan (bottom) of Patient 1. Isotope bone scan fused with lumbar CT scan demonstrates a stress fracture of the left sacral ala (Column 2 = axial, Column 3 = sagittal and Column 4 = coronal). (Column 5) Additional axial images.
the diagnosis without overlap of the pre-isotope diagnostic location in seven of those 14 patients. The major problem with isotope bone scanning in the past has always been poor anatomical resolution of the images. Early studies of isotope scanning for bone-related spinal pathology, using static planar images without SPECT or image fusion, showed that bone scintigraphy failed to identify a number of abnormalities such as pseudoarthrosis [8]. On planar scintigraphy increased activity arising from the posterior vertebral elements is superimposed on
activity from the vertebral body, making localisation of the abnormality difficult [9]. The addition of SPECT improves the utility of the isotope bone scan in detecting pseudarthrosis [10], sources of post-surgery spinal pain [11] and spinal osteomyelitis [12]. Image fusion is the process of image superposition using two or more different sets of images. This process provides the functional information in an anatomical context provided by the CT scan. Typically one anatomical image set, (most commonly CT scan, but
Fig. 2. Patient 2. (Column 1) Coronal CT scan of the cervical spine showing the titanium prosthesis in good position, but linear lucencies between the buttress ends and the adjacent vertebral bone were suggestive of micromovement from non-union. Isotope bone scan and CT scan image fusion shows intense radioisotope uptake in the right C2/3 facet joint (Column 2 = axial, Column 3 = sagittal and Column 4 = coronal).
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034
G.A. Brazenor et al. / Journal of Clinical Neuroscience xxx (2014) xxx–xxx
occasionally MRI) is combined with functional images from conventional nuclear gamma camera SPECT. The first software algorithms for post-scan image fusion required fiducial markers or bands [13,14], but with subsequent algorithms these were no longer required [15,16]. Such imagefusion programs now allow the functional images to be fused with those of high-resolution diagnostic quality CT and MR scanners. Various authors have reported that the improved anatomical resolution aids diagnosis and staging of gynaecological cancers [4], recurrent prostate cancer [5], lymphoma [15], the sensitivity of scanning for pancreatic tumours [17], the diagnosis of hyperparathyroidism [18], and for thyroid disease [19,20]. Using CT scan/ SPECT image fusion in 37 patients with lumbar pain, McDonald et al. reported that image fusion allows definitive localisation of ‘‘hot’’ facet joints, with localisation superior to that of high resolution SPECT alone [21]. In our study cohort the use of high resolution diagnostic CT scan, MRI, or both, was crucial to assess integrity of fusion hardware, disc prolapse, neural compression, and other pathological conditions not readily demonstrated on isotope bone scan alone. The image fusion program permitted functional images to be superimposed on the diagnostic-quality anatomic images from high definition CT scan or MRI. This enabled highly accurate localisation of the scintigraphic changes to anatomic structures such as facet joints, pars fractures and bony fusion masses (both interbody and intertransverse components). In a small number of patients, more distant pathology became evident with the addition of extended imaging fields that were used according to the symptoms of the individual patient. The alternative to image fusion software is scanning with hybrid SPECT/CT cameras. However, the lower cost hybrid SPECT/ CT cameras produce ‘‘non-diagnostic’’ quality CT images [18,22], but nevertheless remain in common use, particularly in oncological investigation [5,18]. Satisfactory image quality can be obtained using the more expensive ‘‘high-end’’ SPECT/CT cameras, but in patients who may already have undergone conventional CT scanning in the investigation of their spinal problem, the subsequent scan using hybrid camera means that the patient receives unnecessary extra radiation. In our study we found that the same anatomical accuracy can be achieved with the use of fusion programs, creating fusion images using the previously acquired anatomical (CT scan or MRI) data set, thus avoiding further irradiation of the patient. 5. Conclusions Investigation of patients presenting with axial spinal pain, limb pain, or both, to a spine surgeon benefits from the addition of Tc99m-MDP isotope bone scan images co-registered with CT scan or MRI. These co-registered images may lead to change in the pathology classification of the provisional diagnosis in 10% of patients, change in the anatomic localisation of the pathology in 15% of patients and major change in treatment intention in 3.6% of patients. We propose that bone scans co-registered with CT scan or MRI are particularly useful in the difficult diagnostic groups where a previous spinal fusion exists or when conventional imaging is normal.
5
Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. References [1] Subramanian G, McAfee JG, Blair RJ, et al. Technetium-99m-methylene diphosphonate – a superior agent for skeletal imaging: comparison with other technetium complexes. J Nucl Med 1975;16:744–55. [2] Cook GJ, Gnanasegaran G, Chua S. Miscellaneous indications in bone scintigraphy: metabolic bone diseases and malignant bone tumors. Semin Nucl Med 2010;40:52–61. [3] Even-Sapir E, Metser U, Mishani E, et al. The detection of bone metastases in patients with high-risk prostate cancer: 99mTc-MDP Planar bone scan, singleand multi-field-of-view SPECT, 18F-Fluoride PET, and 18F-Fluoride PET/CT. J Nucl Med 2006;47:287–97. [4] Tsai CC, Tsai CS, Ng KK, et al. The impact of image fusion in resolving discrepant findings between FDG-PET and MRI/CT in patients with gynaecological cancers. Eur J Nucl Med Mol Imaging 2003;30:1674–83. [5] Schettino CJ, Kramer EL, Noz ME, et al. Impact of fusion of indium111capromab pendetide volume data sets with those from MRI or CT in patients with recurrent prostate cancer. AJR Am J Roentgenol 2004;183:519–24. [6] Amthauer H, Denecke T, Rohlfing T, et al. Value of image fusion using single photon emission computed tomography with integrated low dose computed tomography in comparison with a retrospective voxel-based method in neuroendocrine tumours. Eur Radiol 2005;15:1456–62. [7] Brazenor G. Comparison of multisegment anterior cervical fixation using bone strut graft versus a titanium rod and buttress prosthesis: analysis of outcome with long-term follow-up and interview by independent physician. Spine (Phila Pa 1976) 2007;32:63–71. [8] Hannon KM, Wetta WJ. Failure of technetium bone scanning to detect pseudarthroses in spinal fusion for scoliosis. Clin Orthop Relat Res 1977;123:42–4. [9] Johansen JP, Fossgreen J, Hansen HH. Bone scanning in lumbar disc herniation. Acta Orthop Scand 1980;51:617–20. [10] Gates GF, McDonald RJ. Bone SPECT of the back after lumbar surgery. Clin Nucl Med 1999;24:395–403. [11] Lusins JO, Danielski EF, Goldsmith SJ. Bone SPECT in patients with persistent back pain after lumbar spine surgery. J Nucl Med 1989;30:490–6. [12] Love C, Patel M, Lonner BS, et al. Diagnosing spinal osteomyelitis: a comparison of bone and Ga-67 scintigraphy and magnetic resonance imaging. Clin Nucl Med 2000;25:963–77. [13] Scott AM, Macapinlac H, Zhang J, et al. Image registration of SPECT and CT images using an external fiduciary band and three-dimensional surface fitting in metastatic thyroid cancer. J Nucl Med 1995;36:100–3. [14] Mongioj V, Brusa A, Loi G, et al. Accuracy evaluation of fusion of CT, MR, and Spect images using commercially available software packages (SRS PLATO and IFS). Int J Radiat Oncol Biol Phys 1999;43:227–34. [15] Koral KF, Lin S, Fessler JA, et al. Preliminary results from intensity-based CTSPECT fusion in I-131 anti-B1 monoclonal-antibody therapy of lymphoma. Cancer 1997;80:2538–44. [16] Hutton BF, Braun M, Thurfjell L, et al. Image registration: an essential tool for nuclear medicine. Eur J Nucl Med 2002;29:559–77. [17] Lemke AJ, Niehues SM, Hosten N, et al. Retrospective digital image fusion of multidetector CT and 18F-FDG PET: clinical value in pancreatic lesions-a prospective study with 104 patients. J Nucl Med 2004;45:1279–86. [18] Even-Sapir E, Flusser G, Lerman H, et al. SPECT/Multislice Low-Dose CT: A Clinically Relevant Constituent in the Imaging Algorithm of nononcological Patients Referred for Bone Scintigraphy. J Nucl Med 2007;48:319–24. [19] Profanter C, Wetscher GJ, Gabriel M, et al. CT-MIBI image fusion: a new preoperative localization technique for primary, recurrent, and persistent hyperparathyroidism. Surgery 2004;135:157–62. [20] Lemke AJ, Niehues SM, Amthauer H, et al. Clinical use of digital retrospective image fusion of CT, MRI, FDG-PET and SPECT – fields of indications and results. Rofo 2004;176:1811–8. [21] McDonald M, Cooper R, Wang MY. Use of computed tomography-singlephoton emission computed tomography fusion for diagnosing painful facet arthropathy. Technical note. Neurosurg Focus 2007;22:E2. [22] Nömayr A, Römer W, Strobel D, et al. Anatomical accuracy of hybrid SPECT/ spiral CT in the lower spine. Nucl Med Commun 2006;27:521–8.
Please cite this article in press as: Brazenor GA et al. Co-registration of isotope bone scan with CT scan and MRI in the investigation of spinal pathology. J Clin Neurosci (2014), http://dx.doi.org/10.1016/j.jocn.2013.11.034