Validation of child serum creatinine-based prediction equations for glomerular filtration rate

Area of Science:

• Pediatric Nephrology
• Renal Function Estimation
• Biomarker Accuracy

Background:

• Glomerular filtration rate (GFR) equations are crucial for managing pediatric renal disease.
• The current standard, the Schwartz formula, demonstrates limitations in precision for GFR estimation.
• Accurate GFR assessment is vital for timely diagnosis and treatment of kidney conditions in children.

Purpose of the Study:

• To evaluate the precision and diagnostic capabilities of various pediatric serum creatinine-based GFR prediction equations.
• To compare the performance of the Schwartz formula, Schwartz model, Leger GFR, and Leger model against measured iothalamate GFR (IoGFR).
• To identify the most accurate GFR estimation method for children aged 2-21 years.

Main Methods:

• Retrospective analysis of 195 children (aged 2-21 years) with iothalamate GFR (IoGFR) measurements.
• Estimation of GFR using four equations: Schwartz formula (local k), Schwartz model (regression coefficients), Leger GFR (original coefficients), and Leger model (regression coefficients).
• Evaluation of bias, precision (95% limits of agreement), and diagnostic characteristics (sensitivity, specificity) for each equation.

Main Results:

• All evaluated formulae exhibited poor precision, with 95% limits of agreement around -40 to 40 ml/min per 1.73 m(2).
• Over 72% of GFR estimates fell within 30% of the measured IoGFR across all equations.
• The Schwartz formula showed the highest sensitivity (80%) for detecting GFR <30 ml/min per 1.73 m(2), while the Schwartz model was most sensitive (90%) for GFR <90 ml/min per 1.73 m(2).
• The Leger GFR demonstrated the highest specificity, and the Schwartz model was found to be the most unbiased and sensitive among the models using local coefficients.

Conclusions:

• Existing child serum creatinine-based GFR prediction equations, including the Schwartz and Leger models, lack sufficient precision for clinical use.
• The Schwartz model demonstrated favorable bias and sensitivity characteristics compared to other tested equations.
• Further research is warranted to develop more accurate GFR estimation equations specifically for pediatric populations.