Iterative metal artifact reduction in aortic CTA after Onyx®-embolization

Abstract Purpose Onyx® embolization causes severe artifacts on subsequent CT-examinations, thereby seriously limiting the diagnostic quality. The purpose of this work was to compare the diagnostic quality of the tailored metal artifact reducing algorithms iMAR to standard reconstructions of CTA in patients treated with Onyx® embolization. Method Twelve consecutive patients examined with Dual Energy CTA after Onyx® embolization were included. One standard image dataset without iMAR, and eight image datasets with different iMAR algorithms were reconstructed. Mean attenuation and noise were measured in the aorta or iliac arteries close to the Onyx® glue-cast and compared to the reference level in the diaphragmatic aorta. Mean attenuation and noise were also measured in the psoas muscle close to the Onyx®-glue and compared to the reference level in the psoas muscle at the level of the diaphragm. Subjective image quality and severity of artifacts was assessed by two experienced interventional radiologists blinded to reconstruction details. Results All iMAR reconstructions had less distortion of the attenuation than the standard reconstructions and were also rated significantly better than the standard reconstructions by both interventional radiologists. Conclusion The iMAR algorithms can significantly reduce metal artifacts and improve the diagnostic quality in CTA in patients treated with Onyx® embolization, in many cases restoring non-diagnostic examinations to acceptable diagnostic quality.