Respiratory motion artefact in the liver dome on FDG PET/CT:: comparison of attenuation correction with CT and a caesium external source

Purpose: Respiratory motion has been reported to be a potential cause of artefacts on PET/CT, and of errors in the quantification of lesion activity due to inaccurate attenuation correction. We examined FDG images corrected for attenuation with CT and a caesium external source in the same patients to study this artefact and to assess its impact on detection of lesions in the upper part of the liver. Methods: A total of 122 patients underwent the examination using both attenuation correction techniques, with the Gemini PET/CT scanner. No breathing instructions were given. The images obtained were visually compared, and standardised uptake values (SUVs) in 35 lesions were measured (mean SUV/normal liver SUV) in 14 patients with lesions in the upper part of the liver (less than 5 cm from the upper border). Results: CT-corrected images of the liver included an artefactual cold area in 84 patients (69%); this area was located in the posterior upper part of the liver (65 patients, 53%), included the top of the liver (ten patients, 8%) or affected both the top and the posterior part (nine patients, 8%). In lesions (and also in normal liver outside the artefactual area), SUVs obtained with CT correction were higher than those obtained with Cs correction (p < 0.05), though this was usually without relevance for lesion detection. However, in patients with lesions situated inside the artefactual area, SUVs were lower with CT correction, and ability to detect two lesions (6%) was affected. Conclusion: Failure to detect a liver lesion (especially in the superior and posterior parts) is a rare but possible pitfall when using only CT-corrected FDG images.