Laurence Dubourg1,2,3, Sandrine Lemoine1,2,4, Brune Joannard1, Laurence Chardon5, Vandréa de Souza1,6,7,8, Pierre Cochat2,3,9, Jean Iwaz2,10,11, Muriel Rabilloud2,10,11, Luciano Selistre1,6,7,8. 1. Néphrologie, Dialyse, Hypertension et Exploration Fonctionnelle Rénale, Hôpital Edouard Herriot, Hospices Civils de Lyon, Lyon, France. 2. Université Claude Bernard Lyon 1, Lyon, France. 3. Laboratoire de Biologie Tissulaire et Ingénierie Thérapeutique, CNRS UMR 5305/Université Lyon 1, Lyon, France. 4. Laboratoire CarMeN, INSERM 1060, Lyon, France. 5. Laboratoire de biochimie, Groupement Hospitalier Est, Hospices Civils de Lyon, Lyon, France. 6. Universidade de Caxias do Sul, Programa de Pós-graduação em Ciências da Saúde, Caxias do Sul, Brazil. 7. CAPES Foundation, Ministry of Education of Brazil, Brasilia, Brazil. 8. Hospital Geral de Caxias do Sul, Caxias do Sul, Brasil. 9. Centre de Référence des Maladies Rénales et Phosphocalciques Rares, Hôpital Femme Mère Enfant, Hospices Civils de Lyon, Lyon, France. 10. Service de Biostatistique-Bioinformatique, Pôle Santé Publique, Hospices Civils de Lyon, Lyon, France. 11. CNRS UMR 5558, Laboratoire de Biométrie et Biologie Évolutive, Équipe Biostatistique-Santé, Villeurbanne, France.
Abstract
OBJECTIVES: The one-compartment iohexol plasma clearance has been proposed as a reliable alternative to renal inulin clearance. However, this method's performance depends on the formula used to calculate glomerular filtration rate (GFR). This study reports on performance comparisons between various mathematical formulas proposed for iohexol plasma clearance vs. inulin urinary clearance. METHODS: GFR was simultaneously determined by inulin and iohexol clearance in 144 participants (age: 10-84 years; glomerular filtration rate: 15-169 mL/min/1.73 m2). A retrospective cross-sectional study evaluated the performance of four formulas proposed to calculate plasma iohexol clearance (Brøchner-Mortensen, Fleming et al., Jødal-Brøchner-Mortensen, and Ng-Schwartz-Munoz). The performance of each formula was assessed using bias, precision (standard deviation of the bias), accuracy (percentage iohexol within 5, 10, and 15%), root mean square error, and concordance correlation coefficient vs. renal inulin clearance as reference. RESULTS: Regarding accuracy, there was no difference in root mean square error (RMSE), P5, P10, or P15 between the four formulas. The four concordance correlation coefficients (CCC) between the value from each formula and in-GFR were high and not significantly different. At in-GFR ≥90 mL/min/1.73 m2, Ng-Schwartz-Munoz formula performed slightly better than other formulas regarding median bias (-0.5; 95% CI [-3.0 to 2.0] and accuracy P15 (95.0; 95% CI [88.0-100.0]). CONCLUSIONS: The studied formulas were found equivalent in terms of precision and accuracy, but the Ng-Schwartz-Munoz formula improved the accuracy at higher levels of in-GFR.
OBJECTIVES: The one-compartment iohexol plasma clearance has been proposed as a reliable alternative to renal inulin clearance. However, this method's performance depends on the formula used to calculate glomerular filtration rate (GFR). This study reports on performance comparisons between various mathematical formulas proposed for iohexol plasma clearance vs. inulin urinary clearance. METHODS: GFR was simultaneously determined by inulin and iohexol clearance in 144 participants (age: 10-84 years; glomerular filtration rate: 15-169 mL/min/1.73 m2). A retrospective cross-sectional study evaluated the performance of four formulas proposed to calculate plasma iohexol clearance (Brøchner-Mortensen, Fleming et al., Jødal-Brøchner-Mortensen, and Ng-Schwartz-Munoz). The performance of each formula was assessed using bias, precision (standard deviation of the bias), accuracy (percentage iohexol within 5, 10, and 15%), root mean square error, and concordance correlation coefficient vs. renal inulin clearance as reference. RESULTS: Regarding accuracy, there was no difference in root mean square error (RMSE), P5, P10, or P15 between the four formulas. The four concordance correlation coefficients (CCC) between the value from each formula and in-GFR were high and not significantly different. At in-GFR ≥90 mL/min/1.73 m2, Ng-Schwartz-Munoz formula performed slightly better than other formulas regarding median bias (-0.5; 95% CI [-3.0 to 2.0] and accuracy P15 (95.0; 95% CI [88.0-100.0]). CONCLUSIONS: The studied formulas were found equivalent in terms of precision and accuracy, but the Ng-Schwartz-Munoz formula improved the accuracy at higher levels of in-GFR.
Authors: Pierre Delanaye; Emmanuelle Vidal-Petiot; Thomas Stehlé; Laurence Dubourg; François Gaillard; Gunnar Sterner; Christine A White; Sandrine Lemoine; Vincent Audard; Dominique Prié; Etienne Cavalier; Marie Courbebaisse; Hans Pottel; Martin Flamant Journal: Kidney Int Rep Date: 2021-03-31