The diagnostic value of magnetic resonance imaging in non-traumatic osteonecrosis of the femoral head.

Publisher Website
Google Scholar
Abstract
Radionuclide bone scans, and magnetic resonance-imaging scans. The results of these investigations were compared, for all except one patient, with the pathological findings of transtrochanteric core biopsies of the femoral head and neck of both hips. Of the forty-nine hips, thirty-three had histological proof of osteonecrosis. Twenty-two (67 per cent) of these hips showed definite necrosis on the plain radiographs; eighteen (62 per cent), on the twenty-nine available computed tomographic scans; twenty-four (77 per cent), on the thirty-one available radionuclide bone scans; and all of the hips, on the magnetic resonance-imaging studies. In six additional hips, there were histological changes (marrow necrosis, edema, hemorrhage, and fibrosis) in the medullary spaces without detectable osteonecrosis. The plain radiographs and computed tomographic scans of these six hips were normal except for the computed tomographic scan of one, and the radionuclide uptake on bone-scanning was abnormal in four of the six, as were the magnetic resonance-imaging studies. In the two hips that had normal magnetic resonance-imaging studies, the biopsies showed only destruction of fat cells in the medullary spaces, with no edema or fibroblastic reaction.(ABSTRACT TRUNCATED AT 250 WORDS) To assess the effectiveness of nuclear magnetic-resonance imaging in the detection of osteonecrosis of the femoral head, we studied the cases of twenty-five patients (forty-nine hips) in whom necrosis of the femoral head was suspected on the basis of plain radiographs, computed tomographic scans, radionuclide bone scans, and magnetic resonance-imaging scans. The results of these investigations were compared, for all except one patient, with the pathological findings of transtrochanteric core biopsies of the femoral head and neck of both hips. Of the forty-nine hips, thirty-three had histological proof of osteonecrosis. Twenty-two (67 per cent) of these hips showed definite necrosis on the plain radiographs; eighteen (62 per cent), on the twenty-nine available computed tomographic scans; twenty-four (77 per cent), on the thirty-one available radionuclide bone scans; and all of the hips, on the magnetic resonance-imaging studies. In six additional hips, there were histological changes (marrow necrosis, edema, hemorrhage, and fibrosis) in the medullary spaces without detectable osteonecrosis. The plain radiographs and computed tomographic scans of these six hips were normal except for the computed tomographic scan of one, and the radionuclide uptake on bone-scanning was abnormal in four of the six, as were the magnetic resonance-imaging studies. In the two hips that had normal magnetic resonance-imaging studies, the biopsies showed only destruction of fat cells in the medullary spaces, with no edema or fibroblastic reaction.(ABSTRACT TRUNCATED AT 250 WORDS) Copyright © 1989 by The Journal of Bone and Joint Surgery, Incorporated...