Kinematic MRI of the shoulder.

OBJECTIVE: Clinical evaluation of shoulder pain can be difficult. Pain in the shoulder is often dynamic/positional, and it can often be elicited only during specific activity or function. Soft tissue deforms or changes in shape through motion. These complex interrelationships at the glenoid humeral joint require dynamic studies to accurately evaluate normal anatomy and abnormal pathology. The objective of our study was to evaluate the use of a dynamic cyclic cine imaging to evaluate the glenoid humeral joint. MATERIALS AND METHODS: Kinematic MRI was performed using axial 5 mm sections of the gradient echo and dedicated shoulder surface coil. The patient's arm was placed in the Bonutti cine shoulder system and allowed fixed incremental rotational movement of the shoulder at 10 degrees intervals. Images were reformatted onto a dynamic cine motion by oscillating from internal to external rotation then back to internal rotation. A total of 24 asymptomatic shoulders and 35 symptomatic painful shoulders were studied. RESULTS: Normal variations in the glenoid labrum were readily identified. The glenoid labrum in external rotation is taunt and triangular with well-identified capsular attachments. In neutral rotation the labrum often has increased signal and the middle glenoid humeral ligament occasionally blends with the labrum, making identification difficult. In maximum internal rotation the labrum is rounded and occasionally infolded. Variations in signal through the labrum are not indicative of tears. Capsular attachment must be followed from maximum internal to external rotation to identify stripping. The middle and inferior glenoid humeral ligaments often blend with labrum and internal rotation; however, in external rotation they can be identified as distinct and separate structures. Subcoracoid impingement can be identified in maximum internal rotation with a narrowing of the subcoracoid space to < 11 mm and buckling of the subscapularis and lesser tuberosity against the coracoid process. CONCLUSION: Accurate evaluation of the capsular/labral complex requires cine studies for accurate diagnosis. Signal changes alone or labral morphology alone varies through rotation, and static MRI does not accurately assess these morphologic changes, which vary with extremity position. Capsular attachments can be identified with cine studies accurately and reproducibly. Subcoracoid impingement can be identified with a narrowing of the coracohumeral distance to < 11 mm in internal rotation, which is suggestive of pathology.

The use of computerized tomography in the measurement of glenoid version.

Computerized tomography was done preoperatively on twenty shoulders (thirteen patients) in which there were severe arthritic changes, to measure glenoid version. Ten of the twenty shoulders had osteoarthrosis; eight, rheumatoid arthritis; and two, gouty arthritis. To help determine normal values, computerized tomographic scans of the chest of sixty-three patients who did not have roentgenographic evidence of disease of the shoulder were studied retrospectively for comparison as a control group. In the group of patients who had severe arthritis, the mean glenoid orientation was 11 degrees of retroversion (range, 2 degrees of anteversion to 32 degrees of retroversion). The computerized tomographic scans showed uneven wear of the glenoid surface, osteophytes, large cysts, and posterior displacement of the humeral head. In the control group, the mean orientation of the glenoid was 2 degrees of anteversion (range, 14 degrees of anteversion to 12 degrees of retroversion). The difference between the groups was significant (p less than 0.0001). Glenoid retroversion was increased in the patients who had severe arthritis, and the computerized tomographic scans accurately revealed the extent and pattern of erosion of the bone.

Chest wall mesenchymoma (hamartoma) in infancy. CT and MR findings.

A case of a chest wall mesenchymoma in a five month old infant is presented, and the role of CT and MR are emphasized. There have been no prior reports of the CT or MR findings in this entity.

Early osteonecrosis of the femoral head: detection in high-risk patients with MR imaging.

To determine whether magnetic resonance (MR) imaging can demonstrate the early stages of osteonecrosis that are not detectable radiographically, the authors compared radiologic findings with histologic results in seven patients at high risk for osteonecrosis of the femoral head. Radiography and MR imaging were performed, and proximal femoral intramedullary pressures were measured in all patients, even if results from imaging studies were normal. If the pressures were elevated, core decompression with biopsy was performed. Seven patients had elevated pressures in 11 hips. Of 11 hips from which biopsy specimens were taken, all had histologic evidence of osteonecrosis. However, in only five were the MR imaging findings consistent with osteonecrosis. In the remaining six hips with osteonecrosis, MR imaging findings were normal. Sensitivity of MR imaging in detection of osteonecrosis was 46%. The authors conclude that normal MR imaging results in high-risk patients do not rule out the presence of osteonecrosis.