PT - JOURNAL ARTICLE AU - Laura Perry AU - Bamidele Williams AU - Richard Meades AU - Zarni WIN AU - Kuldip Nijran TI - Identification and reduction of artifacts in dual-isotope infection imaging SPECT/CT: a phantom study DP - 2019 May 01 TA - Journal of Nuclear Medicine PG - 40--40 VI - 60 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/60/supplement_1/40.short 4100 - http://jnm.snmjournals.org/content/60/supplement_1/40.full SO - J Nucl Med2019 May 01; 60 AB - 40Introduction: Interpretation of infection imaging using radiolabelled leukocytes requires understanding of normal and abnormal variants of leukocyte distribution. Dual-isotope studies use Tc-99m colloid to help differentiate marrow distribution from In-111 labelled leukocytes distribution. The addition of Tc-99m colloid can aid diagnosis where bone marrow distribution is disrupted for example following metal joint replacement surgery. Dual isotope infection imaging is included in the SNM [1] and the EANM guidelines [2]. Delayed (1-4hours post injection) and late (16-30 hours post injection) images are acquired. Acquiring the late images with In-111 leukocytes and simultaneously Tc-99m colloid SPECT/CT is time-efficient for the patient and ensures optimal alignment of the datasets. Interpretation of these studies is guided by changes in In-111 leukocyte activity over time and/or identification of regions with In-111 leukocyte activity in the absence of Tc-99m colloid (bone marrow). Artifacts caused by the simultaneous presence of In-111 and Tc-99m and/or metal components could cause lead to false positive or false negative interpretation. Methods: A phantom study was performed to investigate whether artifacts can arise in dual-isotope SPECT/CT images with and without the presence of metal arthroplasty. A phantom was designed and constructed consisting of 3 small sources (In-111 only, Tc-99m only and mixed In-111 and Tc-99m) in a water background. In a second variation of the phantom the mixed In-111 and Tc-99m source was fixed inside a metal resurfacing component. A Siemens Symbia T16 gamma camera was used to acquire SPECT/CT images. Images with and without CT-based attenuation correction were reconstructed using resolution recovery iterative reconstruction (Flash3D) with 8 iterations, 8 subsets and 9mm Gaussian filter. Reconstructions were performed with dual energy window (DEW), triple energy window (TEW) and no scatter correction for the Tc-99m photopeak window. No scatter correction was applied to the In-111 photopeak reconstructions. For each reconstruction the counts in VOIs drawn around each source, for each isotope window, were determined to evaluate any artifacts. Results: In the absence of arthroplasty, artifacts were observed in the Tc-99m images due to cross-talk from the In-111 (up to 13%). The artifacts included increased background activity and appearance of the In-111 only sources. Such artifacts could lead to false-negative interpretations. The magnitude of the artifacts was reduced with scatter correction and TEW reduced the impact of the artifacts to the greatest extent. Artifacts in the In-111 images due to cross-talk from Tc-99m were minimal (<1%). The presence of metal arthoplasty was observed to cause artifacts: increasing the counts in the mixed In-111 and Tc-99m source within the resurfacing component when the CT was used for attenuation correction of the SPECT images. The counts were increased by an average 17% for both isotopes in all reconstructions. The counts were increased by an average 19% if extended CT scale was used within the CT reconstruction. Conclusions: This phantom study has identified artifacts that may occur in dual-isotope infection imaging. The use of TEW scatter correction appears promising to reduce these artifacts. In our centre, work is planned to look at using this in clinical protocols. Interpretation of activity within metal components in SPECT/CT images must be performed with care where CT has been used for attenuation correction by simultaneously reviewing the non-CT attenuation corrected images.