OARSI Clinical Trials Recommendations: Knee imaging in clinical trials in osteoarthritis.

Significant advances have occurred in our understanding of the pathogenesis of knee osteoarthritis (OA) and some recent trials have demonstrated the potential for modification of the disease course. The purpose of this expert opinion, consensus driven exercise is to provide detail on how one might use and apply knee imaging in knee OA trials. It includes information on acquisition methods/techniques (including guidance on positioning for radiography, sequence/protocol recommendations/hardware for magnetic resonance imaging (MRI)); commonly encountered problems (including positioning, hardware and coil failures, sequences artifacts); quality assurance (QA)/control procedures; measurement methods; measurement performance (reliability, responsiveness, validity); recommendations for trials; and research recommendations.

MRI findings of scapholunate instabilities in coronal images: a short communication.

The plain radiographic findings of scapholunate instabilities are very well described in the literature. However, even though the same phenomena that are seen in plain radiographs are also seen in magnetic resonance (MR) images, there are very few reports in the literature regarding MRI findings in scapholunate instabilities. This short communication describes the findings seen in coronal MR images in advanced scapholunate instabilities.

The association of bone marrow lesions with pain in knee osteoarthritis.

BACKGROUND: The cause of pain in osteoarthritis is unknown. Bone has pain fibers, and marrow lesions, which are thought to represent edema, have been noted in osteoarthritis. OBJECTIVE: To determine whether bone marrow lesions on magnetic resonance imaging (MRI) are associated with pain in knee osteoarthritis. DESIGN: Cross-sectional observational study. SETTING: Veterans Affairs Medical Center. PATIENTS: 401 persons (mean age, 66.8 years) with knee osteoarthritis on radiography who were drawn from clinics in the Veterans Administration health care system and from the community. Of these persons, 351 had knee pain and 50 had no knee pain. MEASUREMENTS: Knee radiography and MRI of one knee were performed in all participants. Those with knee pain quantified the severity of their pain. On MRI, coronal T(2)-weighted fat-saturated images were used to score the size of bone marrow lesions, and each knee was characterized as having any lesion or any large lesion. The prevalence of lesions and large lesions in persons with and without knee pain was compared; in participants with knee pain, the presence of lesions was correlated with severity of pain. RESULTS: Bone marrow lesions were found in 272 of 351 (77.5%) persons with painful knees compared with 15 of 50 (30%) persons with no knee pain (P < 0.001). Large lesions were present almost exclusively in persons with knee pain (35.9% vs. 2%; P < 0.001). After adjustment for severity of radiographic disease, effusion, age, and sex, lesions and large lesions remained associated with the occurrence of knee pain. Among persons with knee pain, bone marrow lesions were not associated with pain severity. CONCLUSIONS: Bone marrow lesions on MRI are strongly associated with the presence of pain in knee osteoarthritis.

Interleaved water and fat dual-echo spin echo imaging with intrinsic chemical-shift elimination.

A new technique was developed to simultaneously acquire water and fat dual-echo spin echo images in a single acquisition period. Chemical shifts between water and fat images are intrinsically eliminated, and the images are combined to form water-plus-fat image. In vivo water-only images show fat suppression superior to that of conventional spin echo images. This technique may be clinically useful for musculoskeletal imaging.

3D interleaved water and fat image acquisition with chemical-shift correction.

A new technique, 3D interleaved water and fat image acquisition with chemical-shift correction (3-DIWFAC), was developed to acquire 3D water and fat images in a single acquisition time and to combine the water and fat images to produce chemical-shift-free images. A 3D gradient-recalled-echo (GRE) sequence was implemented with a 1-3-3-1 binomial Shinnar-Le Roux spatial-spectral excitation, and with interleaved phase-encoding lines that alternate between water and fat excitations separated by half TR. Water-only and fat-only images were then realigned to remove chemical shift artifacts. Results from phantoms and human subjects demonstrated that the image contrast was the same as in the regular GRE sequence. With the chemical shift corrected, the shadow artifacts often seen at water and fat boundaries were removed. Since this sequence simultaneously provides water-only images showing cartilage and bone lesions, and water-fat images that depict soft tissue anatomy, it may be clinically useful in musculoskeletal imaging. Magn Reson Med 44:322-330, 2000.

Radiologic measurement of the scapholunate joint: implications of biologic variation in scapholunate joint morphology.

To determine the optimal location for measurement of the scapholunate (SL) joint intercortical width, normal biologic variation in SL joint morphology was evaluated in 40 normal, skeletally mature wrists (16 volunteers, 24 cadavers) using thin-section 1.5T magnetic resonance imaging performed in the axial and coronal planes. The integrity of the SL ligaments was confirmed by magnetic resonance imaging and verified with anatomic dissection of the cadaver wrists. Patterns of SL articular morphology were qualitatively determined using similarity grouping. Scapholunate interval measurements were made at 3 locations each on the mid-SL joint image from both the axial and coronal planes: the articular margins (dorsal-palmar and proximal-distal) and midjoint. Three patterns of midjoint space cortical conformation were observed: parallel congruent (78%), inverted Y (15%), and point-like (8%). The most consistent and narrowest distance between the scaphoid and lunate was found at midjoint: coronal 1.45 mm (44% coefficient of variation) and axial 1.00 mm (22% coefficient of variation). This study demonstrated that measurement of the apparent SL joint interval in an inappropriate site, as with extended or flexed clenched fist views, may provide inaccurate SL joint interval distance assessments. Regardless of SL joint configuration, the midportion of the SL joint shows only moderate biologic variation and the least absolute measurement variance in width and should be the most precise part of the joint to measure. On magnetic resonance imaging, the normal SL joint interval measures less than 2 mm.

A volume adjustable four-coil phased array for high resolution MR imaging of the hip.

A four-coil phased array was specifically designed and built for MR imaging of the hip at 1.5 T. Its RF and imaging properties were evaluated using phantom and in-vivo studies and the results were compared to those of three different commercial coils commonly used for hip imaging. Our coil gave a significantly higher S/N at anatomic locations commonly evaluated for hip diagnosis. The increased S/N supports higher image spatial resolution and improves the visualization of fractures and lateral injuries.

Meniscal subluxation: association with osteoarthritis and joint space narrowing.

OBJECTIVE: Since complete meniscectomy leads to knee OA, we investigated the potential links among meniscal subluxation, joint space narrowing and symptomatic OA. MATERIALS AND METHODS: 233 cases with symptomatic knee OA and 58 asymptomatic controls underwent radiography and MR imaging of the knee. Joint space narrowing was measured on weight-bearing PA fluoroscopy-positioned radiographs. The amount of medial or lateral meniscal subluxation was measured on coronal MR images. The prevalence and severity of meniscal subluxation was compared in cases and controls. We evaluated the correlation of the degree of meniscal subluxation with joint space narrowing, Kellgren and Lawrence grade, and two major risk factors for the development of OA, age and weight. RESULTS: Cases had more medial and lateral subluxation than controls. Mean medial meniscal subluxation was 5.1 mm in cases and 2.8 mm in controls (P=0.001). Modest degrees of meniscal subluxation were common in both cases and controls: 81% of cases and 64% of controls had >/=3 mm of subluxation; age and gender adjusted (P=0.006). Severe degrees of subluxation were almost unique to OA cases (e.g. prevalence of >/=7 mm, 35% cases vs. 7% controls, P< 0.001). Among controls, severe degrees of subluxation were present only in those with radiographic joint space narrowing (defined as >/=grade 1 narrowing on a 0-3 scale). In cases, there was a strong correlation between the degree of medial meniscal subluxation and the severity of medial joint space narrowing (r=0.56, P=0.0001). Similar results were present in the lateral compartment. Meniscal subluxation did not correlate with age or weight. CONCLUSION: Meniscal subluxation is highly associated with symptomatic knee OA. In subjects with osteoarthritis, increasing meniscal subluxation on MR correlates with the severity of joint space narrowing.

Ovarian cancer: comparison of observer performance for four methods of interpreting CT scans.

PURPOSE: To assess the effects of four interpretative methods on observers' mean sensitivity and specificity by using computed tomography (CT) of ovarian carcinoma as a model. MATERIALS AND METHODS: CT scans in 98 patients with ovarian carcinoma and 49 women who were disease free were retrospectively reviewed by four experienced blinded radiologists to compare single-observer reading, single-observer reading with an anatomic checklist, paired-observer reading (simultaneous double reading), and replicated reading (combination of two independent readings). Confidence level scoring was used to identify three possible disease forms in each patient: extranodal tumor, lymphadenopathy, and ascites. Patient conditions were then categorized as abnormal or normal. RESULTS: There were no significant improvements in sensitivity or specificity for classification of patient conditions as abnormal or normal when comparing single-observer interpretation with single-observer interpretation with a checklist or paired-observer interpretation. Although there was no significant improvement in the mean sensitivity (93% vs 94%) by using the replicated reading method, there was a statistically significant improvement in mean specificity (85% vs 79%) for the replicated readings compared with single-observer interpretations (P < .05). CONCLUSION: Diagnostic aids such as checklists and paired simultaneous readings did not lead to an improved mean observer performance for experienced readers. However, an increase in the mean specificity occurred with replicated readings.

MR imaging of rabbit hip cartilage with a clinical imager and specifically designed surface coils.

RATIONALE AND OBJECTIVES: The authors' purpose was to develop a magnetic resonance (MR) imaging technique for examining the hip joint of a rabbit with a clinical MR imager. MATERIALS AND METHODS: Fourteen hips of 10 male New Zealand white rabbits were examined on a 1.5-T MR imager with a specially designed surface coil. Field of view was 3-8 cm, matrix was 256-512 x 192-256, section thickness was 0.9-2.0 mm, and spacing was 0.3-0.5 mm. The hips were sectioned and evaluated by means of light microscopy. Measurements of cartilage thickness from MR images were correlated with those from histologic specimens. RESULTS: The resolution obtainable with a 1.5-T imager was adequate for imaging articular cartilage when a débrided rabbit hip specimen was placed in the center of a 3.5-cm single-loop coil. Rabbit hip cartilage had a trilaminar appearance on MR images. The coefficient of correlation between cartilage-thickness measurements was .81. CONCLUSION: MR imaging of rabbit hip cartilage can be performed on 1.5-T clinical MR imaging unit. Limitations due to the field strength can be overcome with efficiently designed surface coils.

Magnetic resonance imaging: a cost-effective alternative to bone scintigraphy in the evaluation of patients with suspected hip fractures.

OBJECTIVE: To evaluate the cost-effectiveness of magnetic resonance imaging (MRI) compared with radionuclide bone scan in the evaluation of patients with clinically suspected hip fractures. DESIGN: The medical records of all patients who had been seen in the emergency room over a 4 1/2 year period with a clinically suspected hip fracture, negative or equivocal plain films, and either a subsequent bone scan or MRI examination were retrospectively reviewed. The time to diagnosis, admission rate, and time to surgery were determined. A two-sample t-test was used to assess the statistical significance of the results. A theoretical cost analysis was performed using current charges to estimate all expenses. PATIENTS: Forty patients (11 male, 29 female; age 28-99 years) satisfied our inclusion criteria. RESULTS AND CONCLUSIONS: Twenty-one patients had bone scans (six with fractures), and 19 had MRI (four with fractures). The time to diagnosis was 2.24 +/- 1.30 days for bone scanning and 0.368 +/- 0.597 days for MRI (P < 0.0001). Twenty patients in the bone scan group were admitted compared with 13 in the MRI group. The time to surgery was at least 1 day longer in patients undergoing bone scanning. Bone scanning resulted in higher patient costs compared with MRI because of the delay in diagnosis. In the evaluation of patients with suspected hip fractures, early MRI is more cost-effective than delayed bone scanning. Further prospective studies comparing the cost-effectiveness of early MRI with early bone scanning are needed.

Magnetic resonance imaging of the hip with a pelvic phased-array surface coil: a technical note.

OBJECTIVE: The aim of this study was to assess the capability of high-resolution images obtained with a commercially available pelvic phased-array surface coil to demonstrate normal hip anatomy. DESIGN: We retrospectively analyzed the oblique coronal magnetic resonance (MR) images of hips of 36 consecutive patients acquired on a 1.5-T clinical imager using a pelvic phased-array coil as a receiver, a 16-20 cm field of view, and 5 mm slice thickness. PATIENTS: Thirty-six patients were studied, age 15-81 years. There were 20 males and 16 females. RESULTS AND CONCLUSIONS: The articular cartilage, cortex, superior labrum, and iliofemoral ligament were well visualized on proton density weighted fat saturation (PDF) images. The femoral and obturator vessels, obturator nerve, and various muscles were easily seen on T1-weighted images. High-resolution imaging of the hip is achievable in a reasonable amount of time using newer phased-array surface coils and may play an increasing role in the future evaluation of hip disorders.

Scapholunate ligament: normal MR appearance on three-dimensional gradient-recalled-echo images.

PURPOSE: To evaluate the (a) capability of coronal three-dimensional (3D) gradient-recalled-echo (GRE) images in the demonstration of the volar, middle, and dorsal portions of the scapholunate ligament (SLL); (b) normal magnetic resonance (MR) appearance of these portions and of their attachment to the lunate and scaphoid; and (c) normal appearance of the ligament-cartilage interface for various portions of the SLL. MATERIALS AND METHODS: Coronal 3D GRE imaging was used to study the volar, middle, and dorsal portions of the SLL in 14 patients with an arthroscopically normal SLL and in five cadaveric wrists that had a normal SLL proved with dissection. RESULTS: The trapezoidal volar portion of the SLL was seen with inhomogeneous high intermediate signal intensity and attached directly to the lunate and scaphoid cortex. The triangular middle portion was seen with inhomogeneous intermediate signal intensity and in most cases attached to the hyaline cartilage of the lunate and scaphoid. The low-signal-intensity bandlike dorsal portion attached either to the cartilage, cartilage and cortex, or cortex alone of the lunate and scaphoid. CONCLUSION: The volar, middle, and dorsal portions of the SLL can be differentiated on the basis of MR appearance on 3D GRE images. The various portions attach to cartilage and/or to cortex, and the appearance of the ligament-cartilage interface follows a specific pattern.

Lesions of the triangular fibrocartilage complex: MR findings with a three-dimensional gradient-recalled-echo sequence.

PURPOSE: To evaluate the use of a three-dimensional gradient-recalled-echo (GRE) magnetic resonance (MR) imaging sequence in the depiction of lesions of the triangular fibrocartilage (TFC) complex. MATERIALS AND METHODS: MR images of the TFC complex were evaluated in 31 patients who underwent wrist arthroscopy less than 6 months after MR imaging. The results were compared with the arthroscopic findings. RESULTS: Eleven of 12 full-thickness TFC tears were depicted, but one partial-thickness tear, one abnormal disk, and one normal disk were overstaged. Lesions in the volar and dorsal radioulnar ligament and lesions of the attachments of the TFC complex to the ulna often were overstaged. Lesions of the ulnolunate and ulnotriquetral ligaments often were understaged. CONCLUSION: Imaging with the three-dimensional GRE sequence is reliable in the depiction of TFC tears and the exclusion of tears of components of the TFC complex other than those of the ulnolunate and ulnotriquetral ligaments.

Triangular fibrocartilage complex: normal appearance on coronal three-dimensional gradient-recalled-echo MR images.

PURPOSE: To evaluate the usefulness of three-dimensional gradient-recalled-echo (GRE) sequences in demonstration of the triangular fibrocartilage complex (TFCC). MATERIALS AND METHODS: Appearance of the TFCC in wrists of 11 cadavers, 17 patients, and six volunteers at three-dimensional GRE magnetic resonance (MR) imaging was evaluated and compared with that at dissection. RESULTS: MR appearance of the TFCC was similar for all wrists. Volar and dorsal radioulnar ligaments were identified. The TFCC had two types of attachments to the ulna: The more common consisted of two striated attachments, one to the tip and the other to the base of the styloid; the less common consisted of a broad-based striated attachment along the styloid. MR appearance of the ulnotriquetral and ulnolunate ligaments and meniscus homologue varied but correlated with observations at dissection. CONCLUSION: The three-dimensional GRE sequence consistently demonstrated most TFCC components. Dissection helped confirm MR-demonstrable variability in the ulnar aspect of the TFCC.

MR imaging of the triangular fibrocartilage complex.

This article describes in detail the anatomic components of normal triangular fibrocartilage complex (TFCC) and its MR appearance on high resolution MR images. It also describes pathologic processes of TFCC and its MR appearance.

Primary lymphoma of bone. Correlation of magnetic resonance imaging features with cytokine production by tumor cells.

BACKGROUND: Primary lymphoma of bone is a rare, aggressive neoplasm that can present with a large, soft-tissue mass despite minimal evidence of cortical destruction on plain radiographs. METHODS: High resolution magnetic resonance imaging (MRI) examinations of four patients with primary lymphoma of bone were reviewed retrospectively, and in each case intramedullary tumors demonstrated "penetrating channels" extending through the cortex. The MRI studies were correlated with the histopathologic assessment of the tumor for each patient. Immunohistochemistry was performed for immunophenotyping and for cytokine expression by tumor cells. The cytokines that were investigated were interleukin-1, interleukin-6, and tumor necrosis factor-alpha, molecules known to regulate osteoclastic activity. RESULTS: The linear cortical foci noted on MRI correlated with the histopathologic findings of tumor-associated cutting cones, in proximity to osteoclastic bone resorption. Immunohistochemical stains showed a B-cell phenotype for each tumor and positive immunoreactivity in tumor cells for cytokine mediators that stimulate osteoclastic activation. CONCLUSIONS: These findings indicate that the tumor cells in these cases produce soluble cytokine mediators that may regulate extensive osteoclastic activity. In primary lymphoma of bone, tumor activation of osteoclastic resorption, with production of tumor tunnels through the cortex, may represent one of the mechanisms by which lymphoma escapes the intramedullary space and forms large, soft-tissue masses without extensive cortical destruction.