ABSTRACTS PRESENTADOS AL 11th INTERNATIONAL SYMPOSIUM "RADIONUCLIDES IN NEPHRO-UROLOGY", REALIZADO EN MONTEREY, CALIFORNIA, 13-17 MAYO 2001.
31. COMPARISON OF FOUR ALGORITHMS FOR ESTIMATING Tc99mMAG3 CLEARANCE FROM SINGLE BLOOD SAMPLE.
N. Shuke, W. Yamamoto, K. Usui, A. Okizaki, J. Sato, Y. Ishikawa, S. Kaneko, S. Yachiku, K. Kikuchi and T. Aburano.
Asahikawa Medical College, Asahikawa, Japan.
To compare the accuracy in four algorithms for estimating Tc99mMAG3
single sample clearance, seventy-nine patients with a wide range of
renal dysfunction were studied. After bolus injection of Tc99mMAG3
(185 MBq), 8-point blood sampling was performed from 2 to 44 min.
The reference clearance was calculated based on the 2-compartment
model. The algorithms compared were Bubeck's, Russell's ('89), Russell's
('96) and Christensen-Groth's. Based on these algorithms, necessary
constants for the converting formulas were determined by regression
analysis. Using determined constants, single sample clearances were
determined at 20,30 and 44 min after the injection and compared to
the reference. Correlation coefficients 
,
standard errors of estimate (SE) and constants of variation among
sampling times (CV) were summarized in the table. Russell's ('96)
showed the best correlation at every sampling time. In constants of
variation, Russell's ('96) and Christensen-Groth's showed better values
as compared to the other two algorithms. From these results, Russell's
('96) was thought to be the best among these four algorithms.
Correlation
coefficients (SE in 
)
|
|
 |
