reliability of the 123 i- m ibg heart/mediastinum ratio: results of a multicenter test–retest reproducibility study

A quantitative measurement, the heart-to-mediastinum (h/m) ratio of counts derived from a planar acquisition approximately 4 hours after injection of 123i-mibg, is a strong predictor of outcomes in patients with stable class ii-iii heart failure and lvef ≤ 35%. this study assessed the test–retest reproducibility of the h/m ratio in such patients. 47 subjects with class ii-iii systolic heart failure and lvef ≤ 35% were tested at two time intervals separated by 5 to 14 days. subjects were imaged twice on the same camera using the same radionuclide dose. images were sent to a core analysis lab, where three nuclear technologists independently determined the h/m ratios. the primary endpoint was test–retest h/m ratio reproducibility calculated as the absolute difference in mean value determined by the three readers. mean subject age was 65 ± 12 years, 85% were male, and mean bmi was 29 ± 6 kg/m2. mean injected activity was 10.18 ± 0.43 mci for first dose and 10.09 ± 0.52 mci for the second dose. the mean and sd values for first and repeat studies were almost identical: the 95% confidence interval of the mean test–retest difference was 0.055 to 0.076. bland–altman plots showed no systematic effect of the h/m ratio on the magnitude of the difference between replicate measurements. inter-reader measurements were nearly identical. there were no serious adverse events despite exposure to 123i-mibg on 2 occasions in a short time period. the heart-to-mediastinum ratio of 123i-mibg is a consistent and highly reproducible measurement in stable class ii to iii heart failure patients.