Epithelioid hemangioendothelioma of bone.

Neoplasms of bone can arise from any of the cellular elements that constitute osseous tissues. Although tumors of vascular origin are not uncommon, the vast majority are benign. A rare malignant vascular tumor--epithelioid hemangioendothelioma of bone--classifically affects young males and produces osteolytic lesions involving the cortex and cancellous bone of the lower extremities. We present a case with these findings, as well as such unusual findings as cervical spine instability and lesions affecting no fewer than 45 different bones. We conclude that epithelioid hemangioendothelioma should be investigated by skeletal survey because (1) osteolytic lesions involving more that 50% of the cortex present a serious risk for pathologic fracture and (2) the natural history of multicentric epithelioid hemangioendothelioma is more indolent than its solitary counterparts.

Magnetic resonance imaging of the diabetic foot.

MR is useful in evaluating foot abnormalities in diabetic patients. MR offers superior spatial resolution over radionuclide bone scans and superior soft tissue contrast over CT and radiography. MR is able to localize and distinguish soft tissue abscesses from soft tissue edema. MR is sensitive at demonstrating the presence and extent of osteomyelitis.

Gamekeeper thumb: differentiation of nondisplaced and displaced tears of the ulnar collateral ligament with MR imaging. Work in progress.

Ulnar collateral ligament (UCL) injury of the first metacarpophalangeal joint (gamekeeper thumb) is common. If the UCL becomes displaced superficially to the adductor pollicis aponeurosis, surgical treatment has been advocated. Radiography cannot help differentiate between displaced and nondisplaced tears. The authors investigated the use of magnetic resonance (MR) imaging in the evaluation of UCL injury. Ten displaced and six nondisplaced UCL tears were surgically created in 16 nonembalmed cadaveric specimens. Twelve unaltered specimens served as controls. Coronal MR images were obtained and interpreted by two observers. The MR images were compared with corresponding anatomic slices. MR imaging depicted UCL displacement in all 10 specimens with displaced tears. A displaced tear was interpreted in one control specimen. Non-displaced tears were diagnosed in four control specimens. Although MR imaging was only 67% specific for all tears, it was 100% sensitive and 94% specific for depicting UCL displacement and, therefore, may be useful for evaluating gamekeeper thumb.

Magnetic resonance imaging detection of early experimental periostitis. Comparison of magnetic resonance imaging, computed tomography, and plain radiography with histopathologic correlation.

This study characterizes the appearance of periosteal reaction by magnetic resonance imaging (MRI), and evaluates the efficacy of MRI versus computed tomography (CT), and plain film radiography (PF) in detecting early, experimentally induced periostitis. Acute Staphylococcus aureus osteomyelitis was induced in 30 legs of 20 New Zealand white rabbits. The rabbits were then imaged with MR, contrast-unenhanced CT, and PF 4 days after infection. Histologically, periosteal elevation was present in 27 cases. Periosteal ossification was seen in 23 cases, and cellular reaction without ossification in 4 cases. Periosteal reaction was demonstrated by PF in 21 (78%) and by CT in 20 (74%) cases. Evidence of periostitis was seen by MR in all 27% (100%) cases. MR resulted in two false-positive diagnoses. Multiple concentric, alternating high and low signal arcs demonstrated by MR in 19 (70%) cases represented periosteal ossification surrounded by fibrous or granulation tissue. These findings demonstrate the ability of MR to detect periostitis despite the absence of periosteal ossification. MR was more sensitive than CT (P less than .05) or PF (P less than .05) in the detection of experimentally induced periostitis.

THE USE AND UTILITY OF THE MONOTONE CRITERION IN MULTIDIMENSIONAL SCALING.

The monotone criterion as a multidimensional scaling technique is theoret- ically and empirically evaluated using the algorithms of Kruskal (MDSCAL) and Guttman-Lingoes (SSA-1). Geometric configurations are used to test the recovery capabaty and other aspects of MDSCAL and SSA-1. In addition to theoretical shortcomings, the monotone criterion permits results which do not correspond to the shape of the input data. Because of its deficiencies, alterna- tive goodness-of-fit criteria are suggested in preference to the monotone criterion.