The Error of Estimated GFR in Type 2 Diabetes Mellitus

• Main Authors: Sergio Luis-Lima, Tomás Higueras Linares, Laura Henríquez-Gómez, Raquel Alonso-Pescoso, Angeles Jimenez, Asunción María López-Hijazo, Natalia Negrín-Mena, Candelaria Martín, Macarena Sánchez-Gallego, Sara Judith Galindo-Hernández, Raquel Socas Fernández del Castillo, Manuel Castilla-Marrero, Santiago Domínguez-Coello, Vanesa Vilchez de León, Rafael Valcárcel-Lopez, Nerea Insausti-Garmendia, Beatriz Escamilla, Sara Estupiñán, Patricia Delgado-Mallén, Ana-María Armas-Padrón, Domingo Marrero-Miranda, Ana González-Rinne, Rosa María Miquel Rodríguez, María Angeles Cobo-Caso, Laura Díaz-Martín, Federico González-Rinne, Alejandra González-Delgado, Marina López-Martínez, Alejandro Jiménez-Sosa, Armando Torres, Esteban Porrini
• Format: Article
• Language: English
• Published: MDPI AG 2019-09-01
• Series: Journal of Clinical Medicine
• Subjects: type ii diabetes mellitus; renal disease; estimated glomerular filtration rate; measured glomerular filtration rate; plasma clearance of iohexol
• Online Access: https://www.mdpi.com/2077-0383/8/10/1543

Type 2 diabetes mellitus represents 30−50% of the cases of end stage renal disease worldwide. Thus, a correct evaluation of renal function in patients with diabetes is crucial to prevent or ameliorate diabetes-associated kidney disease. The reliability of formulas to estimate renal function is still unclear, in particular, those new equations based on cystatin-C or the combination of creatinine and cystatin-C. We aimed to assess the error of the available formulas to estimate glomerular filtration rate in diabetic patients. We evaluated the error of creatinine and/or cystatin-C based formulas in reflecting real renal function over a wide range of glomerular filtration rate (from advanced chronic kidney disease to hyperfiltration). The error of estimated glomerular filtration rate by any equation was common and wide averaging 30% of real renal function, and larger in patients with measured glomerular filtration rate below 60 mL/min. This led to chronic kidney disease stages misclassification in about 30% of the individuals and failed to detect 25% of the cases with hyperfiltration. Cystatin-C based formulas did not outperform creatinine based equations, and the reliability of more modern algorithms proved to be as poor as older equations. Formulas failed in reflecting renal function in type 2 diabetes mellitus. Caution is needed with the use of these formulas in patients with diabetes, a population at high risk for kidney disease. Whenever possible, the use of a gold standard method to measure renal function is recommended.