The heart-to-mediastinum (H/M) ratio on myocardial scintigraphy with 123I-metaiodobenzylguanidine (MIBG) is used as a semi-quantitative index. However, the scatter from a photopeak of 529 keV on 123I is thought to affect the H/M ratio, and collimator selection is important as well. We attempted to determine the usefulness of low- and medium-energy general purpose (LME) collimators by comparing them with low-energy high-resolution (LEHR) and medium-energy low-penetration (MELP) collimators in phantom and clinical studies. In the phantom study, we used a thoracic phantom and plastic bottles filled with 123I-MIBG solution as upper limbs. Phantom images were acquired with LEHR, LME, and MELP collimators. Regions of interest were placed on the lung, mediastinum, heart, and liver. The average counts in the lung, coefficient of variation (CV%) in the heart, mediastinum, and liver, and H/M ratio were calculated. The H/M ratios obtained with the LEHR collimator and LME collimator were compared in a clinical study. We found that the average count in the lung measured with the LME collimator was reduced to about 30% of that obtained with the LEHR collimator in the phantom study. CV% measured with the LME collimator improved about 10% compared with that determined with the MELP collimator. The H/M ratio measured with the LME collimator was close to that measured with the MELP collimator. In the clinical study, the H/M ratios measured with the LEHR and LME collimators showed a positive relationship (y=2.1x-1.3, x; H/M with LEHR, y; H/M with LME) . LME collimators provided improved contrast and signal-to-noise ratio in evaluation of the H/M ratio on 123I-MIBG myocardial scintigraphy.
