Kidney Week

Abstract: FR-PO0696

Performance of the U25 Estimated Glomerular Filtration Rate (eGFR) Equations in Children with Nephrotic-Range Proteinuria or Primary Nephrotic Syndrome (NS) Diagnoses

Session Information

• Pediatric Nephrology: CKD, ESKD, and Glomerular Diseases
  November 07, 2025 | Location: Exhibit Hall, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Pediatric Nephrology

• 1900 Pediatric Nephrology

Authors

• Dell, Katherine MacRae, Cleveland Clinic Children's Hospital, Cleveland, Ohio, United States

Katherine MacRae Dell, MD

• Reidy, Kimberly J., Children's Hospital at Montefiore, New York, New York, United States

Kimberly J. Reidy, MD

• Seegmiller, Jesse C., University of Minnesota Twin Cities, Minneapolis, Minnesota, United States

Jesse C. Seegmiller, PhD

• Warady, Bradley A., Children's Mercy Kansas City, Kansas City, Missouri, United States

Bradley A. Warady, MD

• Furth, Susan L., The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, United States

Susan L. Furth, MD, PhD

• Ng, Derek K., Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, United States

Derek K. Ng, PhD

Group or Team Name

• CKID Investigators.

Background

Serum creatinine (Cr) and cystatin C (CysC)-based U25eGFR equations developed using data from the Chronic Kidney Disease in Children (CKID) cohort (Pierce C et al, Kidney Int 2021) show good performance relative to measured iohexol GFR (mGFR). However, several studies have highlighted the bias of eGFR equations in patients with proteinuric renal diseases including NS. Performance of eGFR equations has not been rigorously studied in children with CKD and nephrotic-range proteinuria or those with a primary NS diagnosis (dx).

Methods

We analyzed data from 2 CKID cohorts <18 years of age: (1) any CKD diagnosis and nephrotic-range proteinuria (urine protein:creatinine (UPC) ratio ≥ 2.0 mg/mg); and (2) primary NS dx (focal segmental glomerulosclerosis, FSGS or congenital nephrotic syndrome, CNS) with any proteinuria level. U25eGFR was evaluated by bias, correlation, proportion within 10% (P10) and 30% (P30) of mGFR. Observations with complete data on mGFR, Cr and CysC, which did not contribute to the development of the U25eGFR equation, were included. Generalized estimating equations accounted for repeated measures to obtain valid standard errors.

Results

Cohort 1 had 92 participants (126 mGFRs). Mean mGFR=32.3 ml/min/1.73m² (SD: 18.1, mean age=13.5, 38% female). Cr eGFR had a non-significant bias but CysC eGFR (alone or with Cr) overestimated GFR by ~2 ml/min/1.72m² (p<0.001). Cohort 2 had 44 participants (88 mGFRs). Mean mGFR= 51.5 ml/min/1.73m² (SE: 28.4; mean age= 14.8, 40% female). Average biases were not significant for Cr, CysC or combined (Cr+CysC) eGFRs. For both cohorts, Cr + CysC eGFRs provided the highest values for P10 (Cohort 1: 34.9%; Cohort 2: 37.5%), P30 (Cohort 1: 75.4%; Cohort 2 75%), and the strongest correlations (>0.899).

Conclusion

U25eGFR performed reasonably well in this population with nephrotic-range proteinuria or a primary NS dx. There was a significant, although relatively small, overestimation bias in the cystatin equation for those with proteinuria. Overall the combined Cr+Cys C eGFRs for both cohorts provided the highest P10 and P30 with the strongest correlations. Notably, the participants in both cohorts had moderate-severe CKD, so may not reflect eGFR performance in similar patients with higher GFRs.

Funding

• NIDDK Support

Digital Object Identifier (DOI)

• doi: 10.1681/ASN.2025xbeatvxg