Metallic spinal artifacts in magnetic resonance imaging.

STUDY DESIGN

The magnetic resonance artifact susceptibility of traces of surgical aluminum, titanium, and stainless steel in a human spine model was investigated. Metallic filings were deposited in noncontiguous disc spaces in five human thoracic spines before magnetic resonance imaging with spin echo and gradient echo sequences.

OBJECTIVES

Spin echo and gradient echo sequences were used for quantitate and compare void artifact produced by commonly used surgical metals. This was compared to a liquid paraffin control.

SUMMARY OF BACKGROUND DATA

No significant susceptibility artifact was seen with any metal in all spin echo sequences, including T1 (TR 600, TE 12), T2 (TR 2000, TE 30), proton density (TR 2000, TE 80), and fast T2 scanning (TR 3800, TE 96, Ef).

METHODS

Sagittal magnetic resonance imaging permitted void artifact quantification and comparison between different metallic alloys. Two neuroradiologists, working on a blinded basis, evaluated all data and rated the void susceptibility artifact on a scale of 1 (least) to 4 (greatest).

RESULTS

In general, the magnitude of an imaging artifact during magnetic resonance imaging correlated with the magnetism of the metal. Nickel, found in a larger concentration in 316L than in 304 stainless steel, decreases the magnetic resonance artifact of specific metals because of its ability to stabilize iron in a non-magnetic state. Therefore, the 316L stainless steel yielded less artifact production than 304 stainless steel on gradient echo imaging.

CONCLUSION

If upon gradient echo imaging in the postoperative period significant artifact production is noted, stainless steel deposition should be suspected as the causative agent. In this situation, spin echo techniques should be the first approach for attempting optimal visualization of the spinal cord and soft tissue structures.