Hiroyuki Tsushima

Localization of Metastases from Malignant Pheochromocytoma in Patients Undergoing 131I-MIBG Therapy with Manually Fused 123I-MIBG SPECT and CT Images

131I-metaiodobenzylguanidine (MIBG) has been used as a therapeutic agent for pheochromocytoma. Tumor localization and precise staging are essential for therapy with high-dose 131I-MIBG. The sites and extent of 123I-MIBG uptake are usually estimated to predict the effectiveness of therapy before administration. However, conventional scintigraphic images provide insufficient anatomic information. Therefore, we tried to manually superimpose 123I-MIBG SPECT and CT images using free software.

Improved detection of sentinel lymph nodes in SPECT/CT images acquired using a low- to medium-energy general-purpose collimator.

Purpose The use of the low-energy high-resolution (LEHR) collimator for lymphoscintigraphy causes the appearance of star-shaped artifacts at injection sites. The aim of this study was to confirm whether the lower resolution of the low- to medium-energy general-purpose (LMEGP) collimator is compensated by decrease in the degree of septal penetration and the reduction in star-shaped artifacts. Methods A total of 106 female patients with breast cancer, diagnosed by biopsy, were enrolled in this study. 99mTc phytate (37 MBq, 1 mCi) was injected around the tumor, and planar and SPECT/CT images were obtained after 3 to 4 hours. When sentinel lymph nodes (SLNs) could not be identified from planar and SPECT/CT images by using the LEHR collimator, we repeated the study with the LMEGP collimator. Results Planar imaging performed using the LEHR and LEHR + LMEGP collimators positively identified SLNs in 96.2% (102/106) and 99.1% (105/106) of the patients, respectively. Using combination of planar and SPECT/CT imaging with the LEHR and LEHR + LMEGP collimators, SLNs were positively identified in 97.2% (103/106) and 100% (106/106) of the patients, respectively. Conclusions The LMEGP collimator provided better results than the LEHR collimator because of the lower degree of septal penetration. The use of the LMEGP collimator improved SLN detection.

Elimination of scattered gamma rays from injection sites using upper offset energy windows in sentinel lymph node scintigraphy.

ObjectiveThe identification of sentinel lymph nodes (SLNs) near injection sites is difficult because of scattered gamma rays. The purpose of this study was to investigate the optimal energy windows for elimination of scattered gamma rays in order to improve the detection of SLNs. MethodsThe clinical study group consisted of 56 female patients with breast cancer. While the energy was centred at 140 keV with a 20% window for Tc-99m, this energy window was divided into five subwindows with every 4% in planar imaging. Regions of interest were placed on SLNs and the background, and contrast was calculated using a standard equation. The confidence levels of interpretations were evaluated using a five-grade scale. ResultsThe contrast provided by 145.6 keV±2% was the best, followed by 140 keV±2%, 151.2 keV±2%, 134.4 keV±2% and 128.8 keV±2% in that order. When 128.8 keV±2% and 134.4 keV±2% were eliminated from 140 keV±10% (145.6 keV±6%), the contrast of SLNs improved significantly. The confidence levels of interpretation and detection rate provided by the planar images with 140 keV±10% were 4.74±0.58 and 94.8%, respectively, and those provided by 145.6 keV±6% were 4.94±0.20 and 100%. ConclusionBecause lower energy windows contain many scattered gamma rays, upper offset energy windows, which exclude lower energy windows, improve the image contrast of SLNs near injection sites.

Prototype imaging protocols for monitoring the efficacy of iodine-131 ablation in differentiated thyroid cancer.

Whole-body and single photon emission tomography (SPET) images during sodium iodide-131 (Na131I) ablation are useful to confirm the efficacy of ablation using 131I imaging. However, there have been no attempts to improve the quality of 131I imaging. We therefore investigated imaging protocols for 131I imaging in differentiated thyroid cancer (DTC). Phantoms containing 131I were used to simulate extra-thyroid beds and thyroid beds. To simulate extra-thyroid beds, a phantom containing 0.19, 0.37, 0.74 or 1.85 MBq was placed in the acquisition center. To simulate the thyroid beds, four phantoms were applied as normal thyroid tissue, and four phantoms containing 0.19, 0.37, 0.74 and 1.85 MBq were arranged around normal thyroid tissue as a cancer. Whole-body imaging was performed at different table speeds, and SPET data acquired with various pixel sizes were reconstructed using a filtered backed projection (FBP) and ordered-subsets expectation maximization with 3-dimensional (OSEM-3D) algorithm. We measured full width at half maximum (FWHM) and % coefficient of variation (%CV). Patients were then examined based on the results of phantom studies. In extrathyroid beds, slower table speed in whole-body imaging improved %CV, but had little effect on FWHM. For SPET imaging OSEM-3D produced high-resolution and low-noise images, and FWHM and %CV improved with smaller pixel size, as compared with FBP. In the thyroid beds, only the 1.85 MBq phantom could be confirmed on whole-body imaging. Images by SPET had high FWHM and low %CV when the smaller pixel size and OSEM-3D were applied. Accumulation of ≤1.85 MBq was detected with a smaller pixel size of ≤4.8 mm and OSEM-3D. For Na131I ablation imaging, slower scan speed is suitable for whole-body imaging and smaller pixel size and OSEM-3D is appropriate for SPET imaging. In conclusion, we confirmed Na131I accumulation in thyroid beds using slower scan speed (≤15 cm/min) on whole-body imaging, and then accurate identification of Na131I accumulation using SPET and CT fusion imaging with smaller pixel size (≤4.8 mm) and OSEM-3D.