Here is the patient's spine image, in grey-scale, compared to that of another woman with osteoporosis. This should make it easier to notice the abnormalities.

The radiologist totally ignored the dramatic changes seen on the image of the spine. She had a severe inflammation of the dura, requiring extensive surgery that removed posterior elements of her lumbar spine. She told me she mentioned this to the radiologist, but he said it did not affect the bone density reading. This is not true: if there is surgical removal of bone, the projected area will be the same but the mineral content will be much lower.

It is not clear why the report included only L2 - the surgical changes were there, too.

The value at her hip was, therefore, more indicative of her true skeletal status. The radiologist did not ever report the results of the femoral neck - only the Ward's triangle. This region is computer-generated and is not very reliable. It has the worst precision of any area. Her hip BMD showed only osteopenia, and the radiologist did not comment on the huge discrepancy between the spine and hip BMD.

He also said that the change from 1997 to 1999 was "slight" - again, looking only at Ward's triangle. The patient had been placed on estrogen, and her femoral neck showed a 5.6% increase, which is a good increase and unlikely to have occured by chance.

If a physician had only the radiologist report and not the actual scans, she would be completely misled.

Updated 8/8/05