Abstract Archives of the RSNA, 2007
Mohiuddin Hadi MD, Presenter: Nothing to Disclose
Stephen L. Bacharach PhD, Abstract Co-Author: Nothing to Disclose
Millie Whatley, Abstract Co-Author: Nothing to Disclose
V. Koneti Rao, Abstract Co-Author: Nothing to Disclose
Stephen E. Straus, Abstract Co-Author: Nothing to Disclose
Jorge Carrasquillo MD, Abstract Co-Author: Nothing to Disclose
True quantitation of glucose influx (Ki) in 18F-fluorodeoxyglucose(FDG) positron emission tomography(PET) requires an input function. In it’s absence in clinical studies, the Standardized Uptake Value(SUV) is used as a semiquantitative surrogate of Ki. The key assumption of SUV use is that it’s denominator, the ratio of injected dose(d) to either lean body mass (LBM) or body weight(BW), is proportional to the input function integral(AUC). A reference organ whose uptake is consistently dependent primarily on the input function could provide a similar estimate of AUC, with potential advantages and easy implementation clinically. We hypothesized the cerebellum cortex as one such organ. Some investigators have recently postulated the lesion to cerebellum ratio, but did not validate their assumption with actual input functions. Therefore, we decided to compare late static cerebellar cortical counts to AUC.
24 patients(16M, 8F aged 7-46y) with a nonmalignant lymphoproliferative condition underwent 60 minute dynamic FDG PET imaging. Where the index lesion and heart were axially distant, a 25 minute dynamic acquisition over the heart was followed by dynamic imaging of the index lesion for the remaining time. Input functions from 0-60’ post injection were derived from either a basal LV region’s time activity curve(TAC) for 60 minutes, or from a 25 minute TAC interpolated to 60 minutes with additionally drawn blood counts. Using this, AUC was calculated by integration. On the wholebody scan starting at 90 minutes, mean(C) and maximum(Cm) counts in cerebellar cortical regions were obtained. The correlations of C, Cm, d/BW and d/LBM with AUC were examined.
The correlation with AUC was excellent for C (r=0.98) and Cm (r=0.97) , which compared favorably to d/LBM(r=0.95) and d/BW(r=0.98), used in SUVs currently.
Cerebellar cortical counts on a wholebody FDG PET study appear to be a reliable indicator of AUC, comparing favorably to d/LBM or d/BW used in SUV calculations currently. Larger studies in more diverse patient groups, should be performed to validate our findings and its utility for quantitation of clinical FDG PET.
FDG PET Quantitation
Hadi, M,
Bacharach, S,
Whatley, M,
Rao, V,
Straus, S,
Carrasquillo, J,
FDG-PET Uptake in the Cerebellar Cortex May Be a Reliable Indicator of the Input Function Integral. Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL.
http://archive.rsna.org/2007/5012590.html