| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, Pa. From the 2001 RSNA scientific assembly. Received June 23, 2002; revision requested August 22; revision received October 6; accepted December 19. Address correspondence to H.P.L., Radiologisches Institut, Kantonsspital Basel, Petersgraben 4, 4031 Basel, Switzerland (e-mail: hans-peter.ledermann@gmx.ch).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Contrast material–enhanced spinal MR images obtained in 46 consecutive patients (22 women, 24 men; mean age, 58.2 years) with culture or histologic examination results positive for spinal infection were systematically evaluated by two observers. Tuberculous and postoperative infections were excluded. Disk signal intensity and disk height, presence of the nuclear cleft, vertebral signal intensity alterations, endplate erosions on T1-weighted MR images, and presence of paraspinal or epidural inflammation were evaluated. Patient charts and surgical reports were reviewed.
RESULTS: In the 44 patients with disk infection, MR imaging criteria with good to excellent sensitivity included presence of paraspinal or epidural inflammation (n = 43, 97.7% sensitivity), disk enhancement (n = 42, 95.4% sensitivity), hyperintensity or fluid-equivalent disk signal intensity on T2-weighted MR images (n = 41, 93.2% sensitivity), and erosion or destruction of at least one vertebral endplate (n = 37, 84.1% sensitivity). Effacement of the nuclear cleft was only applicable in 18 patients (n = 15, 83.3% sensitivity). Criteria with low sensitivity included decreased height of the intervertebral space (n = 23, 52.3% sensitivity) and disk hypointensity on T1-weighted MR images (n = 13, 29.5% sensitivity). Involvement of several spinal levels occurred in seven (16%) patients. Other spinal infections included isolated vertebral osteomyelitis (n = 1) and primary epidural abscess (n = 1).
CONCLUSION: Most MR imaging criteria commonly used to diagnose disk infections offer good to excellent sensitivity. In atypical manifestations of proven spinal infections, however, some of the classically described MR imaging criteria may not be observed.
© RSNA, 2003
Index terms: Magnetic resonance (MR), contrast enhancement, 30.12143 • Spine, infection, 30.201, 30.22, 30.231 • Spine, intervertebral disks • Spine, MR, 30.121411, 30.121412, 30.121413, 30.121415, 30.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Conditions such as degenerative disk disease with associated endplate edema, inflammatory spondyloarthropathy, hemodialysis associated spondyloarthropathy, neuropathic arthropathy, erosive intervertebral osteochondrosis, and occasionally spinal neoplasms may, however, lead to signal intensity (SI) alterations that may be mistaken for infection (3,6–11).
Several MR imaging patterns and SI alterations have been described to be indicative of spinal infection, including decreased disk height (2–4), disk hypointensity on T1-weighted MR images (1,12), disk hyperintensity on T2-weighted MR images (1,2,4,12,13), disk enhancement (7,12), effacement of the nuclear cleft (1,2), and erosion of the vertebral endplates on T1-weighted MR images (1,4,6,12,14).
It was the goal of this study to systematically evaluate MR imaging findings described as being indicative of spinal infection in patients with proven spinal infection.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Underlying conditions included no associated condition identified (n = 13, 28%), diabetes mellitus (n = 7, 15%), intravenous drug abuse (n = 6, 13%), postoperative nonspinal infection (n = 5, 11%), endocarditis (n = 4, 9%), immunosuppression after liver transplantation or bone marrow transplantation (n = 2, 4%), rheumatoid arthritis (n = 2, 4%), sarcoidosis (n = 2, 4%), and other entities (n = 5, 11%), such as urosepsis, septic phlebitis, lupus erythematodes, multiple sclerosis, and chronic leg ulceration. The study was conducted after approval was obtained from our hospital’s institutional review board to review patient images and medical charts. The board determined that this retrospective study could be conducted without requiring acquisition of signed informed consent from the patient population.
MR Imaging
MR imaging was performed with a 1.5-T superconducting system (Signa; GE Medical Systems, Milwaukee, Wis). A dedicated cervical spine coil (GE Medical Systems) was used for patients who underwent imaging of the cervical spine. Transverse and sagittal MR examinations of the thoracic and lumbar spine were performed with a phased-array coil (CTL; GE Medical Systems).
T1-weighted spin-echo MR images were obtained with one to two signals acquired with a repetition time msec/echo time msec of 400–550/8–12 and a matrix size of 256 x 192 or 256 x 256. T2-weighted MR images were obtained by using a fast spin-echo technique with 2,800–4,000/84–120 (effective), two signals acquired, an echo train length of eight, and a matrix size of 256 x 192 or 265 x 265. Contrast-enhanced T1-weighted MR images were obtained in all patients by using a T1-weighted spin-echo sequence in 39 patients and a fast multiplanar spoiled gradient-recalled-echo technique in seven patients with 200/3, a flip angle of 90°, and a matrix size of 256 x 128 or 256 x 192. Gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) at a dose of 0.1 mmol per kilogram of body weight was used as the intravenous contrast agent. Fat-suppressed T2-weighted MR images were obtained in 18 patients, and fat-suppressed T1-weighted MR images were obtained in 32 patients. For T2-weighted and contrast-enhanced T1-weighted MR sequences, fat suppression was accomplished by using selective presaturation of lipid resonant frequency. Fast spin-echo short inversion time inversion-recovery (STIR) MR images were obtained with 3,000–6,000/20–78 (effective)/150–160 (inversion time), an echo train length of eight, and a matrix size of 256 x 128 or 256 x 192. Fast spin-echo STIR images were available in 12 patients.
Imaging Evaluation
Two musculoskeletal radiologists (M.E.S., J.A.C.) reviewed the MR images in consensus. Images obtained with different sequences in one MR examination were reviewed together as a group. At the level of proven infection, the disk height was graded as normal, less than 50% of the normal disk height, more than 50% of the normal disk height, or increased in comparison to normal adjacent disks. SI of the disk was graded on T1-weighted MR images as isointense, hypointense, or hyperintense in comparison to the SI of adjacent disks. On T2-weighted MR images, SI was graded as hypointense, isointense, hyperintense but less so than fluid, or equivalent to that of fluid in comparison to adjacent disks. On contrast-enhanced images, the disk SI was graded as not enhancing, focally enhancing, rim enhancing, or diffusely enhancing.
Presence or absence of a nuclear cleft in the infected disk on T2-weighted MR images was noted as described in previous reports (1,2,15). A positive nuclear cleft sign was defined as effacement of the hypointense nuclear cleft on T2-weighted MR images in the infected disk with the presence of visible nuclear clefts in the adjacent noninfected disks. A false-negative nuclear cleft sign was defined as preserved nuclear cleft despite proven infection of the disk. The nuclear cleft sign was considered not applicable if the adjacent noninfected disks had no visible nuclear clefts on T2-weighted MR images.
At the level of the infection, SI characteristics of the adjacent vertebral bodies were noted. On T1-weighted MR images, SI of the vertebral bodies was graded as isointense, hypointense, or hyperintense compared with that in adjacent vertebral bodies. On T2-weighted MR images, the SI of the involved vertebral bodies was graded as isointense, hypointense, hyperintense but less so than fluid, or equivalent to that of fluid in comparison to adjacent vertebral bodies. On contrast-enhanced images, the degree of contrast enhancement of the vertebral bodies was graded as diffuse, heterogenous, rimlike, or absent. Extent of abnormal SI in the vertebral bodies was graded as involvement of the entire vertebral body, two-thirds of the vertebral body, or one-third of the vertebral body. The vertebral endplates of infected spinal segments were graded on T1-weighted MR images as intact, eroded (hypointense cortical signal disruption without destruction of the adjacent marrow signal), or destroyed (disruption of the cortical signal and adjacent marrow signal). Destruction of the vertebral bodies was graded as destruction of the entire body (collapse), destruction of two-thirds of the body, or destruction of one-third of the body.
Paraspinous extension of inflammatory soft tissue was measured with a picture archiving and communication system, or PACS, workstation (Canon Medical Systems, Irvine, Calif) in the anteroposterior, lateral, posterior, and craniocaudal directions on contrast-enhanced T1-weighted MR images obtained in each patient. It was also evaluated if posterior extension resulted in an epidural abscess or an epidural phlegmon. If several disks had evidence of infection, the involved levels were noted.
SI alterations indicative of vertebral osteomyelitis were hypointensity of the vertebral bone marrow on T1-weighted MR images, hyperintensity of bone barrow on T2-weighted MR images (1,4,13), and contrast enhancement as described in previous reports (2,16). MR imaging criteria for the diagnosis of an abscess were as follows: iso- or hypointensity compared with muscle tissue on T1-weighted MR images with fluid-equivalent SI on T2-weighted MR images and rim enhancement on contrast-enhanced T1-weighted MR images, as described previously (17,18).
MR imaging criteria for the presence of paraspinous and epidural inflammatory tissue were focal hypointensity on T1-weighted MR images, hyperintensity on T2-weighted MR images, and contrast enhancement (1,2,4,19,20). MR imaging criteria for an epidural phlegmon were hyperintensity on T2-weighted MR images with diffuse contrast enhancement (12,20). MR imaging criteria for an epidural abscess were fluid-equivalent SI on T2-weighted MR images and rim enhancement (12,20).
Review of Clinical Data
Forty-six patients had undergone either percutaneous or surgical biopsy of the spine with histologic analysis and culture performed on the samples after imaging. Patient charts and surgical reports were reviewed by a research fellow (H.P.L.) who was not involved in image analysis. For each patient, it was noted whether antibiotic treatment was administered or surgical intervention was performed after MR imaging. In all patients who underwent surgery, the type of procedure and intraoperative findings were noted. Associated medical conditions were reviewed in patient charts. The pathology and microbiology reports for collected biopsy samples and the results of blood cultures were reviewed. A positive diagnosis of spinal infection was defined as culture growth or characteristic histologic findings of infection in the biopsy samples. These histologic findings were aggregates of inflammatory cells (including neutrophils, lymphocytes, histiocytes, and plasma cells), erosions of trabecular bone, and marrow changes that ranged from loss of normal marrow fat with acute infection to fibrosis and reactive bone formation with chronic disease.
Data Analysis
We calculated the sensitivity for the diagnosis of spondylodiskitis for each of the different MR signs described previously. Additionally, we determined sensitivity values for a combination of two MR signs evaluated together. The following MR criteria were evaluated as a combination of two criteria: hyperintensity on T2-weighted images or fluid-equivalent SI, disk enhancement, nuclear cleft sign, endplate erosion or destruction, and presence of paraspinous inflammatory tissue.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
SI of Infected Disks
The SI characteristics observed in the infected disks in 44 patients with evidence of spondylodiskitis at MR imaging are summarized in Table 1.
|
|
|
|
On T2-weighted MR images, disk isointensity was observed in one patient with surgically proven disk infection and positive cultures for Haemophilus bacteria. The patient had discrete rim enhancement of the disk, both endplates were intact on T1-weighted MR images, and only the superior vertebral body had evidence of involvement of approximately one-third of its volume at MR imaging.
Hypointensity of the infected disks on T2-weighted MR images was seen in two patients who had disk enhancement, erosion or destruction of the adjacent endplates, and evidence of infection of the adjacent vertebral bodies at MR imaging (Fig 2). Both patients underwent surgical treatment.
|
|
|
Presence and distribution of the nuclear cleft sign in the 44 patients with spondylodiskitis is summarized in Table 2. A false-negative nuclear cleft sign was observed in three patients: All three infected disks were contrast enhanced and hyperintense on T2-weighted MR images. Two of these patients had intact adjacent endplates but evidence of infection of the adjacent vertebral bodies. Infection was proven in two patients by means of histologic results and in one patient by means of a positive culture for enterococci. In the cervical and thoracic spine, the nuclear cleft sign was not applicable in most patients. In the lumbar spine, the sign was not applicable in more than one-third of patients.
|
|
|
|
|
|
|
At MR imaging, four patients had evidence of involvement of a disk space superior to the primarily infected level, and four patients had evidence of involvement of a disk space inferior to the infected level (one had both superior and inferior involvement). All seven patients underwent surgical treatment for the spinal infection, and five of these patients underwent diskectomy of the additional disks with evidence of infection at MR imaging.
Sensitivity of MR Imaging Findings
The sensitivity for diagnosis of disk infection was calculated for the following MR findings: decreased disk height, 52.3%; disk hypointensity on T1-weighted MR images compared with that in adjacent normal disks, 29.5%; hyperintensity or fluid-equivalent SI of the disk on T2-weighted MR images, 93.2%; contrast enhancement of the disk, 95.4%; presence of a nuclear cleft sign (if applicable), 83.3%; erosion or destruction of at least one adjacent vertebral endplate, 84.1%; and presence of inflammatory paraspinous or epidural tissue, 97.7%.
The sensitivity values for combinations of two MR criteria evaluated together are summarized in Table 6. Only one combination of two MR criteria with a good sensitivity (>75.0%) resulted in a sensitivity of less than 95%: Combined evaluation of endplate destruction or erosion on T1-weighted MR images and the nuclear cleft sign resulted in a sensitivity of 88.0%.
|
Biopsy Results and Therapy
Of the 46 patients, 14 (30%) underwent computed tomography–guided percutaneous biopsy, and 32 (70%) underwent open surgical biopsy. Histologic analysis of the biopsy samples proved infection in 36 (78%) patients. Culture results proved nontuberculous bacterial infection in 29 (63%) patients. The following bacteria were cultured: S aureus (n = 15), streptococci (n = 4), pseudomonas (n = 3), coagulase-negative staphylococci (n = 2), multiresistant staphylococci (n = 2), Escherichia coli (n = 1), enterococci (n = 1), Haemophilus (n = 1), Klebsiella (n = 1), and pneumococci (n = 1). Two patients had positive culture results for two species of bacteria. Blood cultures were performed in 41 patients and were positive in 25 (61%).
Ten patients with disk infections (including two patients with involvement of only one vertebral body) received intravenous antibiotic treatment after the diagnosis was established. Surgical therapy was performed in the remaining 36 patients. The following procedures were performed: corpectomy and diskectomy (n = 20), diskectomy (n = 7), hemicorpectomy (n = 4), debridement (n = 3), and laminectomy and evacuation of an epidural abscess (n = 2). Surgical spinal stabilization required anterior and posterior fusion in 15 patients, anterior fusion in 14 patients, and posterior fusion in two patients.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Atypical SI alterations have been described in mycobacterial spinal infections (4,7,23), in early pyogenic spinal infections (25), and in patients who abuse intravenous drugs (25). Tuberculous spinal infections may manifest a heterogenous marrow pattern with all MR sequences, relative sparing of the disk, and extensive paraspinous involvement (3,4,23). Tuberculous infection can selectively affect only a part of the vertebral body, especially only the posterior elements, without involving the neighboring disks (3,23,26). Skip lesions, large abscesses, epidural extension, and subligamentous spread are seen more frequently than in other spinal infections (7,16,26,27), and atypical forms of spinal tuberculosis can be indistinguishable from spinal neoplasms such as lymphoma (28). In our study, we excluded patients with tuberculous spinal infections so as to analyze the spectrum of SI alterations of pyogenic spinal infections at MR imaging.
Types of Spinal Infections
Most of our patients had disk infections with classic involvement of two adjacent vertebrae (1,3,13). However, three patients with disk infections demonstrated abnormal SI in only one adjacent vertebral body. SI alterations in only one adjacent vertebral body at MR imaging have been reported in a small number of patients with disk infections in several series (4,12,21,25). The SI alterations were attributed to early stage of infection (21,25), and in both of these studies, follow-up MR images demonstrated subsequent involvement of both adjacent vertebral bodies.
Isolated vertebral body infection (spondylitis) without involvement of adjacent disks, as seen in one of our patients, was described previously as a rare finding (25,29). Isolated infection of a vertebral body is also thought to represent an early manifestation of spinal infection (1,25,29).
SI Alterations in Infected Disks
Disk height is classically described to be decreased in patients with disk infection (2–4,21). In our study, more than one-third of patients with proven disk infection had normal height of the involved disks. More than 10% of our patients had increased height of the intervertebral space due to disk abscesses or apparent increased disk height due to collapse of adjacent vertebral bodies. Decreased height of the involved disk therefore resulted in a low sensitivity of 50.0% for the diagnosis of disk infection. Other investigators have reported normal disk height in infected disks in a minority of patients (4,24), primarily in the setting of early infection (4,24); our results suggest that preserved disk height can be observed in patients with advanced infection, since half of these patients required surgical therapy.
On T1-weighted MR images, typical SI alterations of infected disks have been reported as hypointensity when evaluated with 0.15-T and 0.6-T MR imagers (1,13,21), and it was suggested by others (4) that the lower magnetic field strength may account for this finding. Reports of studies in which 1.5-T MR imagers were used described mostly isointensity of infected disks on T1-weighted MR images (4,24), which concurs with our findings.
On T2-weighted MR images, infected disks are typically considered hyperintense compared with normal adjacent disks (13). Most of our patients had hyperintensity or fluid-equivalent SI on T2-weighted MR images (1,2,4,12). Only three infected disks in our study were not hyperintense. Decreased SI or isointensity of infected disks on T2-weighted MR images was described in numerous previous reports as a rare finding (4,5,12,16,21). It is important to realize that disk iso- or hypointensity on T2-weighted MR images does not exclude disk infection but may instead represent early infection (6). Increased disk SI on T2-weighted MR images is, on the other hand, not specific to septic spondylitis but can also be observed in highly vascularized degenerative disks in erosive intervertebral osteochondritis (6).
After contrast material administration, infected disks almost invariably enhance (16), and only two patients in our study group with proven spondylodiskitis did not have disk enhancement (3,5,12). Lack of enhancement of infected disks was reported to occur rarely (12,16). Enhancement of the disk can increase reader confidence for the diagnosis of infection (16) when there are equivocal findings at nonenhanced MR imaging. Enhancement patterns of infected disks include broad, patchy, linear, thin, or thick enhancement either in the center or peripherally (11,12,16). Most of our patients had rim enhancement of the disks; only one-fifth had diffuse disk enhancement.
Nonvisualization of the nuclear cleft was reported to be indicative of spinal infection (1–3,11,15). The sensitivity of this sign, to our knowledge, has not been evaluated systematically. The main limitation of this sign in our study was that nuclear clefts were not visible in most MR examinations of the cervical and thoracic spine. As was already pointed out (11), the nuclear cleft is rarely visible in the cervical and thoracic spine, and its clinical use is therefore limited. Even in the lumbar spine, more than one-third of our patients did not have nuclear clefts. Furthermore, a preserved nuclear cleft does not exclude spinal infection, since three patients in our group had visible nuclear clefts despite proven infection.
Vertebral Bodies and Paraspinous Tissue
Vertebral bodies adjacent to infected disks are typically described as hypointense on T1-weighted MR images and hyperintense on T2-weighted MR images (1–4,11,13), with SI increase on contrast-enhanced images (11). In our group, there were three patients who had involvement of only the superior vertebral body in proven disk infection and normal SI characteristics in the bone marrow of the lower vertebral body. Discordant SI alterations with normal SI on T1- or T2-weighted MR images in the presence of abnormal SI on the other images were reported previously (4,12,25) but were not observed in our patient group. To improve conspicuity of early bone marrow edema, STIR or fat-saturated proton-density–weighted sequences were advocated instead of fast spin-echo sequences (6,10,22). Abnormal SI in the vertebral bodies was described to usually involve the whole vertebra in disk infections (6,14). Although most of our patients had homogenous involvement of the entire vertebral body on T1-weighted, T2-weighted, and contrast-enhanced MR images, partial involvement of the adjacent vertebral bodies was seen in more than one-third of our patients, and rim enhancement or heterogenous enhancement was seen in more than one-fifth of our patients.
Destruction of the vertebral endplates is considered typical for disk infection (1,2,4,21). Most authors evaluate erosion or destruction of the endplates on T1-weighted MR images (1,2,6,11,12), but others find T2-weighted MR images better suited for evaluation of erosion of the endplates (4). The sensitivity of MR imaging for the diagnosis of endplate erosion or destruction, to our knowledge, has not been evaluated, and different sequences have not been analyzed for optimal results. In our patient group, nearly one-fourth of all endplates had no evidence of erosion or destruction on T1-weighted MR images, and seven patients had intact endplates on both sides of an infected disk. Previous reports have also described preserved endplates in proven disk infection (12,24). Lack of endplate involvement on T1-weighted MR images can therefore not be used as a reliable sign to exclude spinal infection. Pseudosparing of the endplate due to chemical shift artifact can be avoided by means of selection of the phase-encoding plane in the craniocaudal direction (30). It has been stated earlier that destruction and collapse of the vertebral bodies is rarely seen in pyogenic spinal infection (3,26). In our patients, however, more than one-third had evidence of vertebral body destruction at MR imaging, and five patients had frank vertebral collapse with resulting gibbus deformity.
Presence of paraspinous inflammatory tissue is often described in spinal infection (2,4,5,13,19,21), and its presence may help considerably in establishing the diagnosis (4). Our experience concurs with descriptions in previous reports, which state that spinal infection is almost invariably associated with some paraspinous or epidural inflammatory tissue. In our patients with spondylodiskitis, more than 90% had evidence of paraspinous inflammatory tissue, and epidural extension was seen in nearly 90%. Only one patient had neither paraspinous nor epidural involvement. Absence of inflammatory paraspinal or epidural tissue may therefore be a valuable sign to exclude spinal infection.
Involvement of multiple spinal levels was seen in seven of our patients. Although typically described as a feature of tuberculous spondylodiskitis, involvement of adjacent or even distant disks can also be observed in nonmycobacterial infections (1,2,16,19,25).
Some limitations apply to our study. Since we included only patients who had positive percutaneous or surgical biopsy results, a selection of patients with advanced infection may have resulted. This may have led to an overestimation of typical SI findings in spinal infection at MR imaging. This selection of only patients with infection did not allow calculation of specificity values for the discussed MR imaging findings. We cannot exclude a reader bias, since readers were not blinded.
We conclude that criteria with a low sensitivity and limited clinical use include hypointensity of the disk on T1-weighted MR images and decreased height of the intervertebral space. Effacement of the nuclear cleft on T2-weighted MR images is only rarely applicable in the thoracic and cervical spine. Criteria with good to excellent sensitivity include evidence of either paraspinal or epidural inflammatory tissue, contrast enhancement of the disk, hyperintensity or fluid-equivalent SI on T2-weighted MR images, and erosion or destruction of the vertebral endplates on T1-weighted MR images. In atypical manifestations of spinal infections, some classically reported SI alterations considered typical of spinal infection at MR imaging may not be observed. Spinal infection may rarely involve only one vertebral body, a vertebral body and the adjacent disk, or exclusively the epidural space. Pyogenic hematogenous infections frequently involve several spinal levels.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, all authors; study concepts, H.P.L., M.E.S., J.A.C.; study design, M.E.S., W.B.M.; literature research, H.P.L.; clinical studies, M.E.S., J.A.C.; data acquisition, H.P.L.; data analysis/interpretation, M.E.S., J.A.C.; statistical analysis, H.P.L.; manuscript preparation, H.P.L., M.E.S., W.B.M.; manuscript definition of intellectual content, M.E.S., J.A.C.; manuscript editing, H.P.L.; manuscript revision/review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Bierry, F. Jehl, N. Boehm, P. Robert, G. Prevost, J.-L. Dietemann, H. Desal, and S. Kremer Macrophage Activity in Infected Areas of an Experimental Vertebral Osteomyelitis Model: USPIO-enhanced MR Imaging--Feasibility Study Radiology, July 1, 2008; 248(1): 114 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.Y. Hsu, C.W. Yu, M.Z. Wu, B.B. Chen, K.M. Huang, and T.T.F. Shih Unusual Manifestations of Vertebral Osteomyelitis: Intraosseous Lesions Mimicking Metastases AJNR Am. J. Neuroradiol., June 1, 2008; 29(6): 1104 - 1110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Rahme and R. Moussa The Modic Vertebral Endplate and Marrow Changes: Pathologic Significance and Relation to Low Back Pain and Segmental Instability of the Lumbar Spine AJNR Am. J. Neuroradiol., May 1, 2008; 29(5): 838 - 842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.J. Kowalski, K.F. Layton, E.F. Berbari, J.M. Steckelberg, P.M. Huddleston, J.T. Wald, and D.R. Osmon Follow-Up MR Imaging in Patients with Pyogenic Spine Infections: Lack of Correlation with Clinical Features AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 693 - 699. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Grover, V. Dhir, R. Aneja, V. Arya, A. Galle, V. Marwaha, and A. Kumar Severe infections following leflunomide therapy for Rheumatoid Arthritis Rheumatology, July 1, 2006; 45(7): 918 - 920. [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. A. Michel, C. W. A. Pfirrmann, N. Boos, and J. Hodler CT-Guided Core Biopsy of Subchondral Bone and Intervertebral Space in Suspected Spondylodiskitis. Am. J. Roentgenol., April 1, 2006; 186(4): 977 - 980. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
