Kidney Week

Abstract: SA-PO921

Accuracy of CKD-EPI GFR Estimating Equations According to the Difference in eGFR Using Cystatin C vs. Creatinine

Session Information

• CKD: Observational Research and Patient-Oriented Interventions
  November 05, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
  Abstract Time: 10:00 AM - 12:00 PM

Category: CKD (Non-Dialysis)

• 2202 CKD (Non-Dialysis): Clinical‚ Outcomes‚ and Trials

Author

• Wang, Yeli, Harvard University T H Chan School of Public Health, Boston, Massachusetts, United States

Yeli Wang,

Group or Team Name

• Chronic Kidney Disease-Epidemiology Collaboration

Background

Estimated glomerular filtration rate (eGFR) using cystatin C is recommended as a confirmatory test, with eGFRcr-cys more accurate than eGFRcys. A large difference between eGFRcys and eGFRcr (eGFRdiff) likely indicates a substantial divergence from usual in the non-GFR determinants of cystatin C, creatinine or both and a potential large error in either eGFR compared to measured GFR (mGFR). However, it is not known whether eGFRcys or eGFRcr-cys is more correct. We aimed to evaluate the performance of eGFRcr, eGFRcys, and eGFRcr-cys compared to mGFR according to the magnitude of eGFRdiff.

Methods

We assessed the CKD-EPI 2021 eGFRcr and eGFRcr-cys, and 2012 eGFRcys among 4,050 participants from 12 studies included in the CKD-EPI 2021 external validation dataset. eGFRdiff was defined as eGFRcys minus eGFRcr. The negative, reference, and positive eGFRdiff categories were defined as <-15, -15 to <15, and ≥15 ml/min per 1.73 m², respectively. We compared bias (median difference in mGFR minus eGFR), P₃₀ (percentage of eGFR within 30% of mGFR), and concordance between eGFR and mGFR categories (<30, 30-59, 60-89, ≥90 ml/min/1.73m²) according to eGFRdiff categories.

Results

In the overall cohort, mean (SD) GFR, age and BMI were 76.4 (29.6) ml/min/1.73m², 57.0 (17.4) years and 26.9 (5.00) kg/m², respectively. As reported before, eGFRcr-cys in the overall dataset had greater accuracy (higher P₃₀ and greater concordance) than eGFRcr or eGFRcys. In the reference eGFRdiff category, all equations displayed similar performance (small differences in bias, similar P₃₀ and concordance). In both negative and positive eGFRdiff categories, eGFRcr-cys generally had better performance (lesser bias, higher P₃₀ and higher concordance) than eGFRcr and eGFRcys (Table). These results were consistent across subgroups of age, sex, and BMI.

Conclusion

In ambulatory clinical settings eGFRcr-cys is more likely to be correct than either eGFRcr or eGFRcys across sex, age, and BMI groups. Future work should evaluate whether clinician knowledge of non-GFR determinants provides more informative decision making.

Funding

• NIDDK Support