PT  - JOURNAL ARTICLE
AU  - Michael Hofman
AU  - David Binns
AU  - Val Johnston
AU  - Shankar Siva
AU  - Mick Thompson
AU  - Peter Eu
AU  - Marnie Collins
AU  - Rodney J. Hicks
TI  - <sup>68</sup>Ga-EDTA PET/CT Imaging and Plasma Clearance for Glomerular Filtration Rate Quantification: Comparison to Conventional <sup>51</sup>Cr-EDTA
AID  - 10.2967/jnumed.114.147843
DP  - 2015 Mar 01
TA  - Journal of Nuclear Medicine
PG  - 405--409
VI  - 56
IP  - 3
4099  - http://jnm.snmjournals.org/content/56/3/405.short
4100  - http://jnm.snmjournals.org/content/56/3/405.full
SO  - J Nucl Med2015 Mar 01; 56
AB  - Glomerular filtration rate (GFR) can accurately be determined using 51Cr-ethylenediaminetetraacetic acid (EDTA) plasma clearance counting but is time-consuming and requires technical skills and equipment not always available in imaging departments. 68Ga-EDTA can be readily available using an onsite generator, and PET/CT enables both imaging of renal function and accurate camera-based quantitation of clearance of activity from blood and its appearance in the urine. This study aimed to assess agreement between 68Ga-EDTA GFR (68Ga-GFR) and 51Cr-EDTA GFR (51Cr-GFR), using serial plasma sampling and PET imaging. Methods: 68Ga-EDTA and 51Cr-EDTA were injected concurrently in 31 patients. Dynamic PET/CT encompassing the kidneys was acquired for 10 min followed by 3 sequential 3-min multibed step acquisitions from kidneys to bladder. PET quantification was performed using renal activity at 1–2 min (PETinitial), renal excretion at 2–10 min (PETearly), and, subsequently, urinary excretion into the collecting system and bladder (PETlate). Plasma sampling at 2, 3, and 4 h was performed, with 68Ga followed by 51Cr counting after positron decay. The level of agreement for GFR determination was calculated using a Bland–Altman plot and Pearson correlation coefficient (PCC). Results: 51Cr-GFR ranged from 10 to 220 mL/min (mean, 85 mL/min). There was good agreement between 68Ga-GFR and 51Cr-GFR using serial plasma sampling, with a Bland–Altman bias of −14 ± 20 mL/min and a PCC of 0.94 (95% confidence interval, 0.88–0.97). Of the 3 methods used for camera-based quantification, the strongest correlation was for plasma sampling–derived GFR with PETlate (PCC of 0.90; 95% confidence interval, 0.80–0.95). Conclusion: 68Ga-GFR agreed well with 51Cr-GFR for estimation of GFR using serial plasma counting. PET dynamic imaging provides a method to estimate GFR without plasma sampling, with the additional advantage of enabling renal imaging in a single study. Additional validation in a larger cohort is warranted to further assess utility.