Digital Variance Angiography in Lower-Limb Angiography with Metal Implants

PURPOSE: The presence of metal implants may reduce angiographic image quality due to automated beam adjustments. Digital variance angiography (DVA) is reported to be superior to digital subtraction angiography (DSA) with increased contrast-to-noise ratio (CNR) and better image quality. The aim of the study was to evaluate whether DVA could counterbalance the image quality impairment of lower-limb angiographies with metal implants. MATERIALS AND METHODS: From November 2019 to January 2020, 85 raw lower-limb iodine contrast angiograms of 12 patients with metal implants were processed retrospectively with DVA analyses. For objective comparison, CNR of DSA and DVA images was calculated and the ratio CNR(DVA)/CNR(DSA) was determined. Visual image quality was evaluated in a paired comparison and by a five-grade Likert scale by three experienced radiologists. RESULTS: The CNR was calculated and compared in 1252 regions of interest in 37 image pairs containing metal implants. The median ratio of CNR(DVA)/CNR(DSA) was 1.84 with an interquartile range of 1.35–2.32. Paired comparison resulted in 84.5% in favour of DVA with an interrater agreement of 83.2% (Fleiss κ 0.454, p < 0.001). The overall image quality scores for DSA and DVA were 3.64 ± 0.08 and 4.43 ± 0.06, respectively (p < 0.001, Wilcoxon signed-rank test) with consistently higher individual ratings for DVA. CONCLUSION: Our small-sample pilot study shows that DVA provides significantly improved image quality in lower-limb angiography with metal implants, compared to DSA imaging. The improved CNR suggest that this approach could reduce radiation exposure for lower-limb angiography with metal implants. LEVEL OF EVIDENCE: Level 4, case studies