Shine-through artifact due to high-radioactivity bladder and bowel gas in 18F-FDG PET/CT: impact of time-of-flight algorithm and radioactivity concentration of urine in the bladder on the occurrence of the artifacts.

Abstract

Shine-though artifact can appear as regions with falsely increased uptake in the immediate vicinity of large hot sources in 18F-FDG PET/CT. This artifact may adversely affect the assessment of tumor involvement in the regions adjacent to the bladder. The purpose of this study was to evaluate the prevalence of shine-through artifacts in clinical 18F-FDG PET/CT examinations and the factors that can influence their occurrence and extent. PET/CT images were acquired with Discovery PET/CT 690. One hundred six patients who underwent 18F-FDG PET/CT for clinical purposes were retrospectively reviewed. PET images were reconstructed using 3-dimensional ordered-subset expectation maximization with and without time-of-flight (TOF). The shine-through artifact was defined as an erroneous accumulation of 18F-FDG between the bladder and the air region in the intestine without attenuation correction (AC) errors. The maximum standardized uptake value (SUVmax) of the artifact was measured, and the effect of TOF on this artifact was evaluated. The SUVmax in the bladder was compared in patients with and without the artifacts. A phantom containing two spheres simulating bladder and rectal gas was imaged while changing the radioactivity of 18F-FDG solution in the bladder sphere. The relationship between the bladder sphere radioactivity and the SUVmax of the shine-through artifacts was evaluated. The shine-through artifact was more frequently observed on PET images reconstructed without TOF (12/106, 11.3%) as compared to PET images with TOF (8/106, 7.5%, p = 0.046). The SUVmax of the shine-through artifacts was significantly decreased by TOF reconstruction compared to non-TOF reconstruction (4.7 ± 1.7 vs 7.6 ± 3.1, p = 0.0078). The mean SUVmax of urinary bladders in patients with the artifacts was significantly higher than those without the artifacts on non-TOF images (74.9 ± 61.1 vs 46.3 ± 35.2, p = 0.038). In the phantom study, the SUVmax of the shine-through artifact increased as the radioactivity in the bladder-simulating sphere increased. Shine-through artifacts were observed in approximately 10% of clinical 18F-FDG PET/CT examinations. Their magnitude is significantly associated with the radioactivity in the bladder and can be reduced by employing TOF. Recognizing this artifact allows for a more accurate interpretation of 18F-FDG pelvic studies.