Agreement between cystatin-C and creatinine based estimated glomerular filtration rate among Ethiopian children.

This study of 350 Ethiopian children found poor agreement between cystatin C- and creatinine-based estimated glomerular filtration rates, revealing that creatinine-based equations significantly overestimate kidney function compared to cystatin C-based estimates, thus highlighting the need for local validation using gold-standard clearance methods.

The Gist

Imagine your kidneys are the super-efficient filtration plant inside your body. Their main job is to clean your blood, removing waste and keeping everything running smoothly. To check if this plant is working well, doctors usually measure something called the Glomerular Filtration Rate (GFR).

Think of GFR as the "gallons per minute" rating of your kidney's filter. The higher the number, the better the filter is working.

The Problem: Two Different Rulers

The problem is that measuring the actual flow of water through the filter is painful, expensive, and requires a needle (like measuring the exact speed of a car by driving it on a test track). So, doctors use "estimates" based on two different clues found in your blood:

1. Creatinine: A waste product from your muscles. Think of this as dirt from a construction site. The more muscle you have (like a construction crew), the more dirt you produce.
2. Cystatin C: A protein produced by all your cells. Think of this as smoke from a factory. It's produced by everyone, regardless of how much "construction" (muscle) is happening.

For a long time, doctors have used the "dirt" (Creatinine) to guess how well the kidney is working. But recently, many experts started using the "smoke" (Cystatin C) because it might be a more honest clue.

The Study: Checking the Rulers in Ethiopia

This study looked at 350 children in Ethiopia (around 10 years old) to see if these two "rulers" agree with each other. They wanted to know: If we measure the kidney's speed using the "dirt" method and the "smoke" method, do we get the same answer?

The Big Surprise:
The two rulers disagreed wildly.

• The "Dirt" Ruler (Creatinine): Said the kids' kidneys were working super fast (very high numbers). It was like looking at a car and guessing it's doing 100 mph.
• The "Smoke" Ruler (Cystatin C): Said the kidneys were working at a normal, steady pace (lower numbers). It was like the speedometer actually showing 70 mph.

In fact, for about two-thirds of the children, the "dirt" ruler made the kidneys look much better than they actually were. It was like a car mechanic telling you, "Your engine is perfect!" when the engine is actually struggling.

Why Did They Disagree?

The researchers found that the "dirt" (Creatinine) method is tricky because it gets confused by muscle mass.

• If a child has a lot of muscle, they produce more "dirt." The kidney has to work harder to clean it up.
• The old formula assumes that if there's a lot of dirt, the kidney must be working incredibly fast to keep up.
• But in reality, the kidney might just be working normally, and the high "dirt" level is just because the child is muscular or growing.

The "smoke" (Cystatin C) method isn't confused by muscle. It just measures the protein directly, giving a clearer picture of the kidney's actual job.

The Takeaway for Everyday Life

This study is a huge wake-up call for doctors in Ethiopia and other developing regions.

1. We might be lying to ourselves: If we only use the Creatinine test, we might think children have "super-kidneys" when they actually have mild issues. We might miss early signs of kidney trouble because the "dirt" test is too optimistic.
2. The "Gold Standard" is missing: The only way to know for sure is to do a perfect test (like measuring the actual water flow), but that's too painful and expensive for healthy kids.
3. The Solution: The researchers suggest that in places like Ethiopia, we should probably trust the "smoke" test (Cystatin C) more than the "dirt" test (Creatinine) when looking at children. Or, we need to invent new, better formulas that work specifically for the bodies of children in these regions.

In short: The old way of checking kidney health in these children is like using a broken speedometer that always says "fast." The new way (using Cystatin C) is like fixing the gauge to show the real speed, ensuring we don't miss any kids who actually need help.

Technical Summary

1. Problem Statement

Chronic Kidney Disease (CKD) is a growing health burden in sub-Saharan Africa (SSA), potentially originating in childhood. Accurate assessment of Glomerular Filtration Rate (GFR) is critical for early detection and clinical management. However, direct measurement of GFR (e.g., via inulin clearance) is invasive and impractical for large-scale use. Consequently, clinicians rely on estimated GFR (eGFR) derived from endogenous markers: serum creatinine (eGFRcr) and serum cystatin C (eGFRcys).

While high-income studies have noted discrepancies between these two markers, it remains unclear which estimation formulas are valid for children in low-income settings like Ethiopia. There is significant uncertainty regarding whether creatinine-based equations (often influenced by muscle mass) or cystatin C-based equations (less influenced by muscle mass) provide a more accurate reflection of kidney function in this specific population. This study aims to quantify the agreement between eGFRcys and eGFRcr in Ethiopian children and identify factors driving discrepancies.

2. Methodology

• Study Design & Population: A cross-sectional analysis of 350 children (aged 7–12 years, mean age 9.8 years) participating in the 10-year follow-up of the iABC (Infant Anthropometry and Body Composition) birth cohort in Jimma Town, Ethiopia.
• Data Collection:
  • Biomarkers: Fasting serum samples were collected. Creatinine was measured via Jaffe alkaline picrate assay; Cystatin C via enhanced immune turbidimetric assay (both on a Roche Cobas c 702 analyzer).
  • Anthropometry: Height, weight, and body composition (Fat Mass [FM] and Fat-Free Mass [FFM]) were measured using air displacement plethysmography (BOD POD).
  • Clinical: C-reactive protein (CRP) was measured to assess inflammation.
• eGFR Calculation:
  • Creatinine-based: Hoste et al. (2014) formula.
  • Cystatin C-based: Berg et al. (2015) formula.
  • Note: The study also evaluated other formulas (e.g., Schwartz 2009/2012) for sensitivity analysis.
• Statistical Analysis:
  • Agreement: Assessed using Bland-Altman plots and calculated the difference ($eGFR_{diff} = eGFRcys - eGFRcr$).
  • Categorization: $eGFR_{diff}$ was categorized as:
    • Concordant: -15 to <15 mL/min/1.73 m².
    • Higher eGFRcr: < -15 mL/min/1.73 m².
    • Higher eGFRcys: ≥ 15 mL/min/1.73 m².
  • Regression: Multinomial logistic regression was used to identify associations between demographic/anthropometric factors and the level of agreement.

3. Key Contributions

• First Regional Validation: Provides the first comprehensive data on the agreement between cystatin C and creatinine-based eGFR specifically in a cohort of Ethiopian children.
• Formula Comparison: Systematically compares multiple pediatric eGFR formulas, highlighting that the choice of formula drastically alters the classification of kidney function status in this population.
• Identification of Bias: Quantifies the systematic overestimation of kidney function by creatinine-based equations in a low-income setting compared to cystatin C.
• Body Composition Analysis: Investigates the role of fat-free mass and inflammation (CRP) in driving discrepancies between the two markers.

4. Results

• Discrepancy in Distributions: There was a marked variation in eGFR distributions depending on the formula used.
  • Median eGFRcys (Berg 2015): 99.4 mL/min/1.73 m² (IQR: 90.0–114.1).
  • Median eGFRcr (Hoste 2014): 123.2 mL/min/1.73 m² (IQR: 110.3–143.8).
  • Observation: Creatinine-based estimates were significantly higher than cystatin C-based estimates.
• Poor Agreement:
  • Only 27.6% (94/341) of children had concordant results.
  • 64.7% (220/341) had significantly higher eGFRcr (negative $eGFR_{diff}$).
  • 7.6% (26/341) had higher eGFRcys.
  • The overall mean bias was -23.7% (creatinine overestimating relative to cystatin C), with limits of agreement ranging from -79.8% to +33.8%.
• Clinical Implications: If the cystatin C-based Berg 2015 equation is considered the more accurate reference, 27.5% of these "healthy" children would be classified as having reduced kidney function (eGFR < 90 mL/min/1.73 m²), whereas creatinine-based equations classified almost none as such.
• Factors Associated with Discrepancy:
  • Children with higher eGFRcr (negative difference) tended to be older and have higher fat-free mass.
  • Multinomial regression showed a marginally significant association where higher fat-free mass was associated with a lower likelihood of having higher eGFRcys (relative to concordant), suggesting that muscle mass may influence creatinine levels independent of GFR.
• Sensitivity Analysis: The widely used Schwartz 2009 formula (currently used in Ethiopian clinical care) showed poor agreement with cystatin C but good agreement with the Hoste 2014 creatinine formula, suggesting the current clinical standard may also be overestimating GFR.

5. Significance and Conclusion

• Overestimation of Kidney Function: The study concludes that creatinine-based eGFR equations likely overestimate kidney function by up to 30 mL/min/1.73 m² in Ethiopian children compared to cystatin C-based estimates. This could lead to the under-diagnosis of early-stage CKD in a population where renal injury may start in childhood.
• Need for Validation: The authors emphasize that no current pediatric eGFR formula is fully validated for low-income African populations. The large discrepancies suggest that non-GFR determinants (such as muscle mass, diet, or inflammation) affect creatinine and cystatin C differently in this demographic than in the high-income populations where the formulas were derived.
• Recommendations:
  • Clinical reliance on creatinine alone may be insufficient for accurate pediatric kidney assessment in Ethiopia.
  • Future research must prioritize gold-standard validation studies (using Inulin or Iohexol clearance) in African children to develop or calibrate population-specific eGFR equations.
  • Iohexol plasma clearance is recommended as a more feasible alternative to Inulin clearance for future validation studies.

Limitations: The study did not use a gold-standard GFR measurement (due to cost/invasiveness), lacked data on pubertal status, and the cohort consisted of urban children who may not represent the entire Ethiopian population.