Response to Dr. Kanal:

We appreciate Dr. Kanal's comments concerning our letter. The premise that “deciding if one should permit a patient into the bore of an MR scanner should always be considered as a risk-benefit ratio to be assessed on a patient-to-patient basis” is one that all of us should abide by and is emphasized in our letter.

We also agree with Dr. Kanal that we did not employ scientifically sound methods in our report. We lacked controls, follow-up, and the detection of subclinical injury. This is precisely why we chose to submit a Letter to the Editor rather than an original research article. We could not, however, ignore our experience and the experience of our neurosurgical colleagues (verbal communications) who deal with hundreds of gunshot-wound victims with spinal cord injury and paralysis each year. We have never seen any evidence of bullet movement, and more important, neurologic deterioration associated with MR scanning at 1.5 T or below. Even after patients were informed about a potential warm feeling about their spine while in the magnet, we never received any reports of discomfort. Our observations, along with the observations of our colleagues who have scanned similar patients, are consistent in that there was no evidence of neurologic deterioration or pain during or after scanning. Although we cannot attest to the presence or absence of subclinical internal injury, even if we could detect a subclinical adverse event, we are not sure how this would affect our decision to put a particular patient in the MR environment. Certainly, subclinical internal injury can occur within many areas of medical practice from various surgical/interventional procedures to pharmaceutical applications.

Notwithstanding the overwhelming circumstantial evidence that the presence of lead bullets or bullet fragments in or near the spinal canal presents minimal risk to the spinal cord–injured or paralyzed patient, Dr. Kanal's comments are valid. We have, in fact, embarked on a plan in the laboratory to pursue the issues of safety in a laboratory model. In addition, we will establish a prospective protocol with pre- and post-MR evaluation of plain radiographs for detection of bullet movement and a detailed neurologic assessment of the patient before and after MR imaging. Before such studies can be completed, we continue to support the use of MR imaging of patients with retained metallic (lead) ballistic fragments in the region of the spine because we believe that the knowledge we gain outweighs potential risks.