Purpose: The gold standard of measuring glomerular filtration rate (mGFR) is to calculate the urinary clearance of exogenous markers, such as iothalamate (CL_IO), yet such methods can be susceptible to inaccuracies, including errors in urinary and serum sample collection as well as assay errors. An alternative is to estimate GFR (eGFR) using the simplified modification of diet in renal disease (MDRD) equation. These methods were compared and an analysis was conducted to explore sources of variability in renal transplant patients.

Methods: Patients in a phase 3b/4 open-label randomized trial underwent renal transplant and had matched mGFR (CL_IO) and MDRD at one or more time points posttransplant. CL_IO was calculated after a subcutaneous 300-mg injection of nonradioactive iothalamate as the product of urinary concentration x volume of timed urine collection divided by the mean of 2 serum concentrations. Comparisons between MDRD and CL_IO used Pearson correlation coefficient and Bland-Altman measure of agreement. Other sources of variability on CL_IO, including the volume of urine and serum concentration of iothalamate, and deviations in the sample collection time of urine or serum, were assessed.

Results: Time-matched CL_IO and MDRD were available for 105 patients at baseline and 42 patients 12 months postrandomization. Baseline mean±SD/median (range) were: CL_IO 73.8 ±103/50.8 (1.75–864) mL/min; MDRD 57.9 ±14.2/53.2 (37.4–96.2) mL/min. Twelve-month postrandomization means ±SD/median (range) were: CL_IO 62.4 ±51.3/49.7 (2.80–232) mL/min; MDRD 63.2 ±12.2/60.5 (39.7–93.6) mL/min. The distribution of CL_IO values was wider than expected. The Bland-Altman analysis revealed a systematic trend between the difference and mean of eGFR and mGFR, indicating a proportional bias. One of the sources of variability in the mGFR assessments included unreasonable volumes of urine collected in the protocol-specified time, with median (range) of 150 mL (5–1340 mL) and median (range) urinary concentrations of 24 mg (0.83–550 mg).

Conclusion: Results indicate considerable variability in CL_IO and lack of concordance compared with MDRD due to errors that included volume collection and assay performance and may call into question results of CL_IO studies.

Tortorici, M.: Employee, Pfizer, Stockholder, Pfizer. Li, H.: Employee, Pfizer, Stockholder, Pfizer. Hahn, C.: Employee, Pfizer, Stockholder, Pfizer. Flynn, A.: Employee, Pfizer. Korth-Bradley, J.: Employee, Pfizer, Stockholder, Pfizer. Schulman, S.: Employee, Pfizer, Stockholder, Pfizer.

« Back to 2013 American Transplant Congress