Magnetic resonance imaging versus radionuclide scintigraphy in screening for bone metastases.

AIM: To determine the frequency of isolated, asymptomatic metastases of the peripheral skeleton in patients with breast and prostate cancer. METHODS: A retrospective review was performed of staging skeletal scintigrams. Patients with hot spots in the peripheral skeleton in the absence of convincing evidence of metastatic disease in the axial skeleton were followed-up to determine the nature of the peripheral lesion(s) based on other imaging modalities, serial imaging or biopsy, and to determine if the lesion(s) had been symptomatic at the time of the scintigram. SUBJECTS: 200 patients with histologically proven carcinomas of the breast or prostate. RESULTS: Four patients (2%) had isolated metastatic involvement of the peripheral skeleton, and would, therefore, have been wrongly staged by a magnetic resonance 'marrow screen' of the axial skeleton. However, in three of these patients the lesions were painful, prompting diagnostic plain radiographs independent of the scintigraphic findings. Twelve patients (6%) had isolated scintigraphic abnormalities of the peripheral skeleton suggestive of metastatic disease but which on further investigation were shown to be benign lesions. CONCLUSION: A limited magnetic resonance (MR) 'marrow screen' confined to the axial skeleton would not result in any significant loss of accuracy in staging patients with breast and prostate carcinoma compared with skeletal scintigraphy. Given the proven increased sensitivity of MR over skeletal scintigraphy in the detection of bone metastases, and the additional information MR provides, it is likely to provide a more accurate basis for management.