A Comparison of the Accuracy of Different Single Plasma Sample Methods for Measuring Glomerular Filtration Rate Using ⁵¹Cr-EDTA in Children

Abstract

Purpose

Among the different methods of measuring glomerular filtration rate (GFR) using ⁵¹Cr-ethylenediaminetetraacetic acid clearance, the two-plasma-sample method (TPSM) is widely used, and highly accurate. The single-plasma-sample method (SPSM) is occasionally used for simplicity, at the expense of accuracy. Our aims were (1) to investigate the correlation and (2) to compare the accuracy of six known SPSMs in pediatric patients in reference to TPSM.

Methods

We retrospectively reviewed 122 pediatric cases (65 boys, age 7.3 ± 4.6 years) and analyzed 307 GFR measurements. SPSMs included Groth and Aasted at 120 min, Ham at 120 min, Christensen and Groth at 120 and 240 min, and Jacobsson at 120 and 240 min. Reference GFR (GFR_ref) was defined using TPSM GFR corrected by the Jodal and Brochner-Mortensen equation. GFR_ref < 30 mL min⁻¹ 1.73 m⁻² were excluded. The standard error of the estimate (SEE) and the number of cases with differences > 10% (N_10%) were used to evaluate accuracy.

Results

SPSMs generally correlated well with GFR_ref (r = 0.92~0.99) and were relatively accurate (SEE = 9.21~15.60). Groth and Aasted showed the smallest SEE, while Jacobsson at 240 min showed the smallest N_10% for all GFR_ref ranges. As for the decreased GFR_ref, Ham was most accurate followed by Jacobsson at 240 min.

Conclusions

Jacobsson at 240 min provided good accuracy in all GFR_ref ranges and was well correlated with TPSM. Jacobsson at 240 min might be the most appropriate method to substitute for TPSM in pediatric patients. Ham could be an alternative in patients with impaired renal function.

Appendices

Appendix 1: Slope-Intercept GFR Corrected Using the Jodal and Bochner-Mortensen Equation (Reference GFR) [26]

$$ {\displaystyle \begin{array}{c}\mathrm{Reference}\ \mathrm{GFR}\ \left(\mathrm{mL}\ {\min}^{-1}\ 1.73\ {\mathrm{m}}^{-2}\right)=\mathrm{SI}-{\mathrm{GFR}}_{\mathrm{cor}}/\left(1+0.00185\times {\mathrm{BSA}}^{-0.3}\times \mathrm{SI}-{\mathrm{GFR}}_{\mathrm{cor}}\right),\mathrm{SI}-{\mathrm{GFR}}_{\mathrm{cor}}\ \left(\mathrm{mL}\ {\min}^{-1}\ 1.73\ {\mathrm{m}}^{-2}\right)=\mathrm{SI}-\mathrm{GFR}\times \left(1.73\ \mathrm{BSA}\ {\mathrm{m}}^{-2}\right)\\ {}\kern0ex \mathrm{SI}-\mathrm{GFR}\ \left(\mathrm{mL}\ {\min}^{-1}\right)=\left({w}_{\mathrm{i}}\times \mathrm{s}\times 200/{w}_{\mathrm{s}}\right)\times \left(\ln \left({C}_1/{C}_2\right)/\mathrm{d}t\right)\times \left({C_2}^{\left(t1/\mathrm{d}t\right)}/{{\mathrm{C}}_1}^{\left(t2/\mathrm{d}t\right)}\right)\end{array}} $$

where w_i is injection mass (g), w_s is standard mass (g), t₁ is 1st plasma sample time (min), t₂ is 2nd plasma sample time (min), dt = t₂ – t₁ (min), s is standard count (cpm mL⁻¹), C₁ is 1st plasma sample count (cpm mL⁻¹), C₂ is 2nd plasma sample count (cpm mL⁻¹), and BSA is body surface area (m²).

Appendix 2: Groth and Aasted’s Method [20]

$$ \mathrm{Cl}\ \left(\mathrm{mL}\ {\min}^{-1}\ 1.73\ {\mathrm{m}}^{-2}\right)=A+\mathrm{BX},A=-553.124\ \ln (t)+3236.76,B=72.295\ \ln (t)-425.41,X=\ln \left\{C(t)\times \mathrm{BSA}\times {10}^7/\mathrm{ID}\right\} $$

where t is sample time (90~120 min), C(t) is sample activity at time = t (min) (cpm mL^-1), ID is injected dose (cpm).

Reference method of plasma clearance: two-exponential curve fitting five samples at 5, 15, 60, 90, and 120 min after the injection of ⁵¹Cr-EDTA.

Appendix 3: Ham’s Method [6, 10]

$$ \mathrm{Cl}\ \left(\mathrm{mL}\ {\min}^{-1}\right)=2.602{V}_{120}-0.273,{V}_{120}\ \left(\mathrm{L}\right)=\mathrm{ID}/{C}_{120},{C}_{120}=A(t)\times {\mathrm{e}}^{(0.008)\left(t-120\right)} $$

where t is sample time (100~130 min), A(t) is actual radioactivity at sample time = t (cpm mL^-1), C₁₂₀ is radioactivity corresponding to sample time = t (cpm mL^-1), and V₁₂₀ is volume of distribution at sample time = t (L).

Reference method of plasma clearance: one-exponential curve fitting two samples between 2 and 4 h after the injection of ⁵¹Cr-EDTA with composite correction constant of 0.85

Appendix 4: Christensen and Groth’s Method [6, 21, 35]

$$ \mathrm{Cl}\ \left(\mathrm{mL}\ {\min}^{-1}\right)=\left(-b+\sqrt{\left({b}^2-4 ac\right)}\right)/2\mathrm{a},a=0.0000017{t}^2-0.0012t,b=-0.000775{t}^2+1.31t,c=\mathrm{ECV}\times \ln \left(\mathrm{ECV}/V(t)\right) $$

where ECV is extracellular volume = 8116.6 BSA − 28.2 (mL), V(t) is volume of distribution = ID/C(t) (mL), t is sample time (180~300 min), BSA is body surface area (m²), and ID is injected dose (cpm).

Reference method of plasma clearance: two-exponential curve fitting 16 samples from 0 to 300 min after the injection of ^99mTc-DTPA.

Appendix 5: Jacobsson’s Method [22]

$$ \mathrm{Cl}\ \left(\mathrm{mL}\ {\min}^{-1}\right)=\ln \left(V(t)/{V}^{\prime}\right)/\left(t/{V}^{\prime }+0.0016\right) $$

where V(t) is volume of distribution at sample time = t (mL), ID is injected dose (cpm), C(t) is plasma radioactivity at sample time = t (cpm mL^-1), V′ = 0.246 BW, BW is body weight (g), t is sample time (min).

Reference method of plasma clearance: one-exponential curve fitting four samples from 240 to 300 min after the injection of ^99mTc-DTPA with Brochner-Mortensen’s correction.