The GFR and GFR decline cannot be accurately estimated in type 2 diabetics

Flavio Gaspari, Piero Ruggenenti, Esteban Porrini, Nicola Motterlini, Antonio Cannata, Fabiola Carrara, Alejandro Jiménez Sosa, Claudia Cella, Silvia Ferrari, Nadia Stucchi, Aneliya Parvanova, Ilian Iliev, Roberto Trevisan, Antonio Bossi, Jelka Zaletel, Giuseppe Remuzzi

Research output: Contribution to journalArticle

94 Citations (Scopus)

Abstract

There are no adequate studies that have formally tested the performance of different estimating formulas in patients with type 2 diabetes both with and without overt nephropathy. Here we evaluated the agreement between baseline GFRs, GFR changes at month 6, and long-term GFR decline measured by iohexol plasma clearance or estimated by 15 creatinine-based formulas in 600 type 2 diabetics followed for a median of 4.0 years. Ninety patients were hyperfiltering. The number of those identified by estimation formulas ranged from 0 to 24:58 were not identified by any formula. Baseline GFR was significantly underestimated and a 6-month GFR reduction was missed in hyperfiltering patients. Long-term GFR decline was also underestimated by all formulas in the whole study group and in hyper-, normo-, and hypofiltering patients considered separately. Five formulas generated positive slopes in hyperfiltering patients. Baseline concordance correlation coefficients and total deviation indexes ranged from 32.1% to 92.6% and from 0.21 to 0.53, respectively. Concordance correlation coefficients between estimated and measured long-term GFR decline ranged from -0.21 to 0.35. The agreement between estimated and measured values was also poor within each subgroup considered separately. Thus, our study questions the use of any estimation formula to identify hyperfiltering patients and monitor renal disease progression and response to treatment in type 2 diabetics without overt nephropathy.

Original languageEnglish
Pages (from-to)164-173
Number of pages10
JournalKidney International
Volume84
Issue number1
DOIs
Publication statusPublished - Jul 2013

Fingerprint

Iohexol
Type 2 Diabetes Mellitus
Disease Progression
Creatinine
Kidney
Therapeutics

ASJC Scopus subject areas

  • Nephrology

Cite this

The GFR and GFR decline cannot be accurately estimated in type 2 diabetics. / Gaspari, Flavio; Ruggenenti, Piero; Porrini, Esteban; Motterlini, Nicola; Cannata, Antonio; Carrara, Fabiola; Jiménez Sosa, Alejandro; Cella, Claudia; Ferrari, Silvia; Stucchi, Nadia; Parvanova, Aneliya; Iliev, Ilian; Trevisan, Roberto; Bossi, Antonio; Zaletel, Jelka; Remuzzi, Giuseppe.

In: Kidney International, Vol. 84, No. 1, 07.2013, p. 164-173.

Research output: Contribution to journalArticle

Gaspari, F, Ruggenenti, P, Porrini, E, Motterlini, N, Cannata, A, Carrara, F, Jiménez Sosa, A, Cella, C, Ferrari, S, Stucchi, N, Parvanova, A, Iliev, I, Trevisan, R, Bossi, A, Zaletel, J & Remuzzi, G 2013, 'The GFR and GFR decline cannot be accurately estimated in type 2 diabetics', Kidney International, vol. 84, no. 1, pp. 164-173. https://doi.org/10.1038/ki.2013.47
Gaspari F, Ruggenenti P, Porrini E, Motterlini N, Cannata A, Carrara F et al. The GFR and GFR decline cannot be accurately estimated in type 2 diabetics. Kidney International. 2013 Jul;84(1):164-173. https://doi.org/10.1038/ki.2013.47
Gaspari, Flavio ; Ruggenenti, Piero ; Porrini, Esteban ; Motterlini, Nicola ; Cannata, Antonio ; Carrara, Fabiola ; Jiménez Sosa, Alejandro ; Cella, Claudia ; Ferrari, Silvia ; Stucchi, Nadia ; Parvanova, Aneliya ; Iliev, Ilian ; Trevisan, Roberto ; Bossi, Antonio ; Zaletel, Jelka ; Remuzzi, Giuseppe. / The GFR and GFR decline cannot be accurately estimated in type 2 diabetics. In: Kidney International. 2013 ; Vol. 84, No. 1. pp. 164-173.
@article{bf2a65fe20af4a96aa442c82a1d2832c,
title = "The GFR and GFR decline cannot be accurately estimated in type 2 diabetics",
abstract = "There are no adequate studies that have formally tested the performance of different estimating formulas in patients with type 2 diabetes both with and without overt nephropathy. Here we evaluated the agreement between baseline GFRs, GFR changes at month 6, and long-term GFR decline measured by iohexol plasma clearance or estimated by 15 creatinine-based formulas in 600 type 2 diabetics followed for a median of 4.0 years. Ninety patients were hyperfiltering. The number of those identified by estimation formulas ranged from 0 to 24:58 were not identified by any formula. Baseline GFR was significantly underestimated and a 6-month GFR reduction was missed in hyperfiltering patients. Long-term GFR decline was also underestimated by all formulas in the whole study group and in hyper-, normo-, and hypofiltering patients considered separately. Five formulas generated positive slopes in hyperfiltering patients. Baseline concordance correlation coefficients and total deviation indexes ranged from 32.1{\%} to 92.6{\%} and from 0.21 to 0.53, respectively. Concordance correlation coefficients between estimated and measured long-term GFR decline ranged from -0.21 to 0.35. The agreement between estimated and measured values was also poor within each subgroup considered separately. Thus, our study questions the use of any estimation formula to identify hyperfiltering patients and monitor renal disease progression and response to treatment in type 2 diabetics without overt nephropathy.",
author = "Flavio Gaspari and Piero Ruggenenti and Esteban Porrini and Nicola Motterlini and Antonio Cannata and Fabiola Carrara and {Jim{\'e}nez Sosa}, Alejandro and Claudia Cella and Silvia Ferrari and Nadia Stucchi and Aneliya Parvanova and Ilian Iliev and Roberto Trevisan and Antonio Bossi and Jelka Zaletel and Giuseppe Remuzzi",
year = "2013",
month = "7",
doi = "10.1038/ki.2013.47",
language = "English",
volume = "84",
pages = "164--173",
journal = "Kidney International",
issn = "0085-2538",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - The GFR and GFR decline cannot be accurately estimated in type 2 diabetics

AU - Gaspari, Flavio

AU - Ruggenenti, Piero

AU - Porrini, Esteban

AU - Motterlini, Nicola

AU - Cannata, Antonio

AU - Carrara, Fabiola

AU - Jiménez Sosa, Alejandro

AU - Cella, Claudia

AU - Ferrari, Silvia

AU - Stucchi, Nadia

AU - Parvanova, Aneliya

AU - Iliev, Ilian

AU - Trevisan, Roberto

AU - Bossi, Antonio

AU - Zaletel, Jelka

AU - Remuzzi, Giuseppe

PY - 2013/7

Y1 - 2013/7

N2 - There are no adequate studies that have formally tested the performance of different estimating formulas in patients with type 2 diabetes both with and without overt nephropathy. Here we evaluated the agreement between baseline GFRs, GFR changes at month 6, and long-term GFR decline measured by iohexol plasma clearance or estimated by 15 creatinine-based formulas in 600 type 2 diabetics followed for a median of 4.0 years. Ninety patients were hyperfiltering. The number of those identified by estimation formulas ranged from 0 to 24:58 were not identified by any formula. Baseline GFR was significantly underestimated and a 6-month GFR reduction was missed in hyperfiltering patients. Long-term GFR decline was also underestimated by all formulas in the whole study group and in hyper-, normo-, and hypofiltering patients considered separately. Five formulas generated positive slopes in hyperfiltering patients. Baseline concordance correlation coefficients and total deviation indexes ranged from 32.1% to 92.6% and from 0.21 to 0.53, respectively. Concordance correlation coefficients between estimated and measured long-term GFR decline ranged from -0.21 to 0.35. The agreement between estimated and measured values was also poor within each subgroup considered separately. Thus, our study questions the use of any estimation formula to identify hyperfiltering patients and monitor renal disease progression and response to treatment in type 2 diabetics without overt nephropathy.

AB - There are no adequate studies that have formally tested the performance of different estimating formulas in patients with type 2 diabetes both with and without overt nephropathy. Here we evaluated the agreement between baseline GFRs, GFR changes at month 6, and long-term GFR decline measured by iohexol plasma clearance or estimated by 15 creatinine-based formulas in 600 type 2 diabetics followed for a median of 4.0 years. Ninety patients were hyperfiltering. The number of those identified by estimation formulas ranged from 0 to 24:58 were not identified by any formula. Baseline GFR was significantly underestimated and a 6-month GFR reduction was missed in hyperfiltering patients. Long-term GFR decline was also underestimated by all formulas in the whole study group and in hyper-, normo-, and hypofiltering patients considered separately. Five formulas generated positive slopes in hyperfiltering patients. Baseline concordance correlation coefficients and total deviation indexes ranged from 32.1% to 92.6% and from 0.21 to 0.53, respectively. Concordance correlation coefficients between estimated and measured long-term GFR decline ranged from -0.21 to 0.35. The agreement between estimated and measured values was also poor within each subgroup considered separately. Thus, our study questions the use of any estimation formula to identify hyperfiltering patients and monitor renal disease progression and response to treatment in type 2 diabetics without overt nephropathy.

UR - http://www.scopus.com/inward/record.url?scp=84879687867&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84879687867&partnerID=8YFLogxK

U2 - 10.1038/ki.2013.47

DO - 10.1038/ki.2013.47

M3 - Article

C2 - 23447062

AN - SCOPUS:84879687867

VL - 84

SP - 164

EP - 173

JO - Kidney International

JF - Kidney International

SN - 0085-2538

IS - 1

ER -