Performance of creatinine-based equations for estimating glomerular filtration rate changes over time

Area of Science:

• Nephrology
• Clinical Chemistry
• Biostatistics

Background:

• Glomerular filtration rate (GFR) is a key marker for monitoring chronic kidney disease (CKD).
• The accuracy of creatinine-based equations for estimating GFR slope over time needs comprehensive evaluation.
• Assessing GFR slope performance across patient demographics and CKD types is crucial.

Purpose of the Study:

• To evaluate the performance of creatinine-based equations (CKD-EPI and MDRD) in estimating GFR slope.
• To determine how patient characteristics and CKD diagnoses affect the accuracy of GFR slope estimation.
• To assess the clinical utility of eGFR slopes for tracking kidney function changes.

Main Methods:

• Utilized the NephroTest cohort study data with 5324 GFR measurements from 1955 adult CKD patients (Stages 1-4).
• Measured GFR using chromium 51-labeled ethylenediamine tetraacetic acid renal clearance.
• Simultaneously estimated GFR using CKD-EPI and MDRD equations and calculated absolute/relative GFR slopes via linear mixed models.

Main Results:

• Mean absolute GFR decline was 1.6 mL/min/1.73 m²/year (mGFR), 1.5 (CKD-EPI), and 1.3 (MDRD).
• Mean relative GFR decline was 5.9%/year (mGFR), 5.3 (CKD-EPI), and 4.8 (MDRD).
• CKD-EPI and MDRD slopes showed poor precision (52-55% within 30% of mGFR) and age-dependent bias, overestimating in younger and underestimating in older patients.

Conclusions:

• Creatinine-based GFR slope estimation exhibits limited bias but poor precision.
• Age-related bias in GFR slope estimation is significant, potentially linked to muscle mass variations.
• Findings highlight the need for caution when using eGFR slopes for clinical decisions and research, especially concerning age.