Glomerular filtration rate in children: validation and longitudinal use of determination from the 99mTc DTPA renogram

The glomerular filtration rate (GFR) was estimated from the technetium diethylenetriaminepentacetic (99mTc DTPA) renogram by empirically correlating the renogram and clearance of the tracer as measured by plasma disappearance kinetics. To expand the potential application of this technique from its original description, minor changes in methodology were made; a new regression was generated with 30 subjects, and that formula was then validated prospectively. For the 50 children and adolescents in the prospective series whose GFR normalized for surface area (NGFR = GFR X 1.73 m2/SA; SA, surface area) was less than 120 mL/min/1.73 m2; renal function was estimated with a mean residual error of 8.0 mL/min/1.73 m2, and a mean relative error of 11%. These results are at least as good as those in the initial report, thus validating this method in a prospective series. To examine the use of this method longitudinally, 129 examinations were reviewed in 55 subjects who had NGFR estimated from the renogram (estimated) on at least two occasions. For 28 sets of serial examinations, NGFR was also measured by the plasma disappearance of 99mTc DTPA (measured). The true change in function for each set of serial examinations was defined as the increment for whichever method (estimation or measurement) was more stable. The mean longitudinal errors were 2.1 and 4.8 mL/min/1.73 m2 for measured and estimated NGFR, respectively. The increments in NGFR for these methods were correlated with a regression line near identity and r = .852. This indicates that a change in estimated NGFR reflects a change in measured NGFR of similar magnitude. This method can accurately estimate GFR from the 99mTc DTPA renogram, and it can be used longitudinally to follow children with renal disorders.