Recent trends in the management of bone metastases include the use of prophylactic bisphosphonates and low dose single fraction radiotherapy in favour of high dose fractionated radiotherapy. A recent animal model [1] suggests that cord compression is often associated with predominant soft tissue epidural disease, with bone collapse as a late event. In the present study, potential implications were investigated by retrospective evaluation of the pattern of disease on MRI scans of patients with spinal cord compression. The dominant component of spinal cord compression was determined in 62 patients. Two main patterns were identified, 45 (73%) had predominant soft tissue epidural disease and 15 (24%) had bone collapse. There were two with intrathecal deposits. The patterns were correlated with response to radiotherapy. Positive response was observed in 64% of those with soft tissue epidural disease and 27% of those with bone collapse. These data support the animal model, suggesting soft tissue epidural disease rather than collapse as the predominant cause of cord compression. This implies that prophylactic bisphosphonates alone would be unlikely to reduce the incidence of spinal cord compression. It also introduces concern about the long term safety of low dose single fraction radiotherapy for bone metastases in patients with a medium term life expectancy (e.g. > 2 years). These patients may benefit from more than a single fraction of radiotherapy to produce longer growth delay for sub-clinical epidural disease.