RESUMO
The aim of this study is to evaluate how metallic artifacts in the lumbar spine can affect images obtained from magnetic resonance (MR) sequences. We performed a phantom experiment by scanning an agar containing an orthopedic metallic implant using 64-channel multidetector row computed tomography (CT) and a 3-tesla MR unit. We compared the reproducibility in each measurement, enlargement or reduction ratio of the CT and MR measurements, and signal deviation in each voxel from the control. The reproducibility on CT and multiacquisition variable-resonance image combination selective (MAVRIC SL) was good, but that on the other MR sequences showed either fixed bias or proportional bias. The reduction ratios of the distance between the nails were significantly smaller in MAVRIC SL than in the other MR sequences after CT measurements (p<0.001, respectively). MAVRIC SL was able to reduce the metallic artifact, permitting observation of the tissue surrounding the metal with good reproducibility.
