# Performance of glomerular filtration rate equations using serum creatinine in children with congenital anomalies of the kidney and urinary tract

## TL;DR

This study compares different methods for estimating kidney function in children with urinary tract anomalies, finding that two formulas are reliable for this purpose.

## Contribution

The study evaluates and compares the performance of three GFR estimation equations in children with CAKUT, a specific pediatric population.

## Key findings

- The bedside Schwartz and EKFC equations showed strong correlation with measured GFR in children with CAKUT.
- The EKFC equation demonstrated higher accuracy in children over two years old compared to the other formulas.
- Both equations are suggested as reliable tools for GFR estimation in this population.

## Abstract

Congenital anomalies of the kidney and urinary tract (CAKUT) are a significant cause of pediatric morbidity and mortality, often leading to chronic kidney disease (CKD). Accurate glomerular filtration rate (GFR) assessment is crucial for effective management, but a gold-standard pediatric GFR estimation formula remains elusive. This study compared the bedside Schwartz, Chronic Kidney Disease in Children (CKiD-U25) creatinine, and European Kidney Function Consortium (EKFC) creatinine-based equations against measured GFR (mGFR) using technetium-99m-diethylenetriaminepentaacetic acid (Tc-99mDTPA) clearance.

Data were collected from 276 children with CAKUT at Dr. Soetomo Academic General Hospital. Estimated GFR (eGFR) was calculated using the bedside Schwartz, CKiD-U25, and EKFC equations. mGFR was determined using Tc-99mDTPA clearance, considered the gold standard. Correlation (Spearman rs), bias, and accuracy (P30, percentage of eGFR within 30% of mGFR) were assessed. Subgroup analysis was performed for children older than two years.

Both equations correlated significantly with mGFR (bedside Schwartz: r = 0.793; CKiD-U25: r = 0,793; EKFC: r = 0.745, p < 0.0001). However, the bedside Schwartz systematically underestimated mGFR (bias: −21.7 mL/min/1.73 m²), while the EKFC overestimated (bias: 17 mL/min/1.73 m², p < 0.0001). In children >2 years, correlations strengthened (bedside Schwartz: r = 0.804; EKFC: r = 0.835, p < 0.0001), with the EKFC demonstrating more accuracy (P30: 90.1%) compared to the bedside Schwartz CKiD-U25 creatinine (75% and 75%, respectively)

These findings suggest the both bedside Schwartz and EKFC could be a reliable tool for GFR estimation in this pediatric CAKUT population.