Estimating and tracking renal function in children and adults with spina bifida

Abstract

Purpose: Estimated glomerular filtration rate (eGFR) in the general population is stable in children after 2 years of age until adulthood. In the first three decades after age 18, eGFR decreases by 0.3-0.8 ml/min/1.73 m²/year. Little data exists regarding eGFR changes in the spina bifida (SB) population given variability in muscle mass. In the absence of a validated SB-specific eGFR formula, the performance of different eGFR formulas may vary. We performed a cross-sectional study (1) to determine trends in eGFR with increasing age in children and adults with SB and (2) to compare eGFRs calculated using different formulas.

Methods: We retrospectively reviewed records of patients 2-50 years old with SB followed at our institution (2014-2019). We determined eGFR using four pediatric formulas (2-17 years: CKiD_SCr, CKiD_Cys, CKiD_SCr-Cys, Zappitelli_SCr-Cys) and four adult formulas (18 + years: MDRD_SCr, CKD-EPI_SCr, CKD-EPI_Cys, CKD-EPI_SCr-Cys). One eGFR per patient was included (most recent eGFR for those with serial measurements). Patients were categorized as chronic kidney disease (CKD) stage 2 (eGFR 60-89) and Stage 3+ (<60). Non-parametric tests, linear regression, and Spearman's correlation were used for analysis.

Results: Among 209 children with SB (median age 10.3 years), depending on the formula used, eGFR decreased by -0.7 to -1.8 ml/min/1.73 m²/year (CKiD_Cys, CKiD_Scr, p ≤ 0.001), remained stable (CKiD_SCr-Cys, p = 0.41), or increased by +2.7/year (Zappitelli_SCr-Cys, p < 0.001) (Figure). The proportion of children with CKD 2 or higher varied between formulas (11.5-58.9%, p < 0.001). Correlations between pediatric formulas were negligible to moderate. Comparing any two formulas, 12.0-65.6% of children were assigned a different CKD stage. Among 164 adults (median age 26.3), eGFR decreased for each formula (range: -1.3 to -2.2/year, p ≤ 0.01) (Figure). The proportion of adults with CKD 2 or higher varied between formulas (9.2-30.5%, p < 0.001). Correlations between adult formulas were moderate to very high. Comparing any two formulas, 8.5-26.8% of adults were assigned a different CKD stage.

Comment: We cannot reliably determine whether eGFR changed during childhood. Among adults, eGFR decreased with age for every formula evaluated at greater than twice the general population rate. Without a validated SB-specific eGFR formula, we are left with formulas providing different results.

Conclusions: Estimated GFR among adults with SB appears to deteriorate at a higher rate than in the general population. Due to lack of precision of accepted eGFR formulas in the SB population, this may be a real phenomenon, an artifact of inaccurate eGFR formulas, or both. A validated SB-specific eGFR formula is needed.

Keywords: Chronic; Creatinine; Cystatin C; Renal insufficiency; Spinal dysraphism.