Intense FDG Focus Adjacent to a Hip Prosthesis on CT Attenuation-Corrected PET
A 71-year-old man with a history of bilateral total hip arthroplasty and a 2.1-mm-thick cobalt-chromium femoral stem undergoes FDG PET/CT for staging of a newly diagnosed melanoma on the left calf. The PET/CT is performed with a low-dose CT for attenuation correction. The reconstructed PET images show a focus of intense uptake (SUVmax 9.6) adjacent to the right femoral prosthesis, and the CT shows marked streak artifact in the same region. The patient has no hip pain, and plain radiographs from 2 months earlier show no loosening. The PET/CT scanner uses time-of-flight and point-spread-function reconstruction. A diagnostic contrast-enhanced CT is not planned. The referring oncologist asks whether this represents a new metastasis.
1. How can CT-based attenuation correction produce apparent FDG uptake near metal implants?
Read the worked answer
CT-based attenuation correction converts the CT Hounsfield units into attenuation coefficients at 511 keV. Metal produces severe streak and beam-hardening artifacts, so the calculated attenuation map near the prosthesis may be incorrect. Areas of soft tissue may be assigned inappropriately high or low attenuation values, and the PET emission data may be over- or under-corrected. The result can be a false focus of high apparent activity adjacent to the implant. Misregistration between the CT and PET can worsen this because the wrong attenuation values are applied to the emission data. Time-of-flight and point-spread-function reconstruction improve image quality but do not eliminate attenuation-correction artifacts from metal.
2. What steps would you take before calling this focus a metastasis?
Read the worked answer
First, review the non-attenuation-corrected PET images. A focus that is present only on the attenuation-corrected images and not on the non-AC images is more likely an artifact. Second, inspect the PET/CT registration and the CT artifact pattern; if the focus aligns with the densest streak, artifact is likely. Third, apply metal-artifact-reduction software if available and compare reconstructed images. Fourth, compare with prior imaging and with the plain radiographs, which showed no loosening. Fifth, correlate clinically: the patient has no hip pain. If the finding remains suspicious, recommend MRI, ultrasound, or biopsy rather than treating it as metastatic disease on PET/CT alone. The final report should state that the focus is indeterminate because of metal artifact.