Estimating Kidney Function for Pediatric Dose Decisions

Why the Kidney-Function Number Must Match the Dose Rule

A pediatric kidney-function estimate is not, by itself, a dose recommendation. It is an input to a rule that may specify an equation, a body-size convention, a renal-function threshold, and a resulting dose or interval. If an estimate is calculated with the wrong equation, reported in the wrong units, or compared with a threshold intended for a different measure, a child can be assigned to the wrong dose category even when the arithmetic appears correct.

This tutorial focuses on estimating kidney function for dose decisions, not on deriving a drug-specific renal dose adjustment. The bedside Schwartz equation is a commonly used pediatric estimate of glomerular filtration rate (GFR). Its result is indexed to a standard body-surface area (BSA). Some dosing rules instead require an absolute value in mL/min, a measured GFR, or another specified estimate. The applicable drug labeling or clinical protocol determines which measure to use.

Definition: The estimated glomerular filtration rate (eGFR) is an equation-based estimate of GFR derived from patient characteristics and a filtration marker, such as serum creatinine. The bedside Schwartz eGFR is conventionally reported in mL/min/1.73 m², meaning that it is scaled to a reference BSA of 1.73 m². An absolute, body-size-adjusted estimate is reported in mL/min.

The Bedside Schwartz Estimate and Its Conditions

For children with a serum creatinine measured in mg/dL, the bedside Schwartz equation is

\mathrm{eGFR}_{\mathrm{Schwartz}} = \frac{0.413 \times \text{height (cm)}}{\text{serum creatinine (mg/dL)}} \quad \text{mL/min/1.73 m}^2.

The coefficient 0.413 is associated with a standardized creatinine assay. The equation is intended for pediatric use; it should not be assumed to be an appropriate equation for every age, clinical state, or dosing decision. It assumes that serum creatinine is sufficiently stable to represent the patient's usual creatinine concentration. A rapidly changing creatinine concentration can lag behind changing filtration, so a single eGFR may not describe current kidney function well.

Height and serum creatinine must be measured accurately and expressed in the units required by the equation. The estimate can also be affected by factors that make serum creatinine an atypical marker of filtration, including unusually low or high muscle mass, poor growth, or severe illness. If the clinical question or drug protocol requires a different validated equation or a measured GFR, the bedside Schwartz result should not silently replace it.

Key interpretation: A bedside Schwartz value of 80 mL/min/1.73 m² is an estimate scaled to 1.73 m². It is not automatically an absolute filtration value of 80 mL/min, and it is not automatically the renal-function variable required by a drug's dose table.

Worked Example: Calculating an Indexed Pediatric eGFR

A hypothetical child is 120 cm tall and has a stable serum creatinine of 0.60 mg/dL measured with a standardized assay. Calculate the bedside Schwartz eGFR.

The required inputs are present: height is in centimeters, creatinine is in mg/dL, and the stated stable value supports interpreting the estimate as a snapshot of usual kidney function.

Substitute the values:

\mathrm{eGFR} = \frac{0.413 \times 120}{0.60} = \frac{49.56}{0.60} = 82.6\ \text{mL/min/1.73 m}^2.

The result is 82.6 mL/min/1.73 m². It is an estimate, not a direct measurement. Whether 82.6 falls into a particular dosing category depends on the threshold and kidney-function measure specified by the relevant dosing rule.

Equation Choice Can Change a Dose Category

Pediatric equations are developed and validated in particular populations and with particular laboratory methods. A coefficient from an older equation cannot be substituted into a newer equation just because both use height divided by serum creatinine. For example, an older Schwartz form used a coefficient of 0.55 for many children, while the bedside form above uses 0.413 with standardized creatinine. The values are not interchangeable: assay calibration and the population used to develop an equation matter.

The following comparison is a sensitivity illustration, not a recommendation to calculate both equations and select whichever produces a preferred dose. Assume a hypothetical protocol explicitly calls for a particular estimate and uses a boundary at 75 mL/min/1.73 m². The example shows why equation selection must be resolved before applying that boundary.

Worked Example: An Equation Difference Near a Dose Boundary

A hypothetical 10-year-old girl is 140 cm tall, has a stable standardized serum creatinine of 0.80 mg/dL, and is being considered under a hypothetical protocol that assigns the full dose at eGFR values of at least 75 mL/min/1.73 m² and a reduced dose below 75. Calculate the two estimates for comparison.

Using the bedside Schwartz coefficient:

\mathrm{eGFR}_{\mathrm{bedside}} = \frac{0.413 \times 140}{0.80} = \frac{57.82}{0.80} = 72.3\ \text{mL/min/1.73 m}^2.

Using the older coefficient of 0.55, purely to illustrate the effect of changing equations:

\mathrm{eGFR}_{\mathrm{older}} = \frac{0.55 \times 140}{0.80} = \frac{77.0}{0.80} = 96.3\ \text{mL/min/1.73 m}^2.

Under the hypothetical boundary, 72.3 is below 75 and 96.3 is above it. The resulting category changes with the equation. This does not establish which estimate is clinically correct for a real patient: that depends on the equation specified and validated for the intended use. It demonstrates why the equation and assay convention must be fixed before categorization, particularly near a threshold.

Indexed and Absolute Values Are Different Quantities

An indexed eGFR can be converted to an absolute estimate when a protocol requires mL/min. First calculate BSA, for example with the Mosteller equation, using height in centimeters and weight in kilograms:

\mathrm{BSA\ (m^2)} = \sqrt{\frac{\text{height (cm)} \times \text{weight (kg)}}{3600}}.

Then de-index the eGFR by multiplying by the patient's BSA and dividing by 1.73 m². This conversion changes the body-size scaling; it does not improve the underlying estimate or turn eGFR into a measured GFR.

Conversion: If the indexed eGFR is in mL/min/1.73 m², the corresponding absolute estimate is $$ \mathrm{eGFR}_{\mathrm{absolute}} = \mathrm{eGFR}_{\mathrm{indexed}} \times \frac{\mathrm{BSA\ (m^2)}}{1.73\ \mathrm{m^2}}, $$ reported in mL/min.

Worked Example: De-Indexing Before Applying an Absolute Threshold

A hypothetical child weighs 24 kg and is 120 cm tall. The indexed eGFR calculated earlier is 82.6 mL/min/1.73 m². A hypothetical dosing rule uses an absolute threshold of 50 mL/min. Calculate BSA and the absolute estimate before assigning a category.

Calculate BSA:

\mathrm{BSA} = \sqrt{\frac{120 \times 24}{3600}} = \sqrt{0.8000} = 0.8944\ \mathrm{m^2}.

De-index the eGFR:

\mathrm{eGFR}_{\mathrm{absolute}} = 82.6 \times \frac{0.8944}{1.73} = 42.7\ \text{mL/min}.

The indexed estimate is 82.6 mL/min/1.73 m², whereas the absolute estimate is 42.7 mL/min. Under the hypothetical rule using an absolute threshold of 50 mL/min, the absolute estimate is below the boundary. Comparing the indexed value of 82.6 directly with that absolute threshold would mix units and could assign the wrong category.

Creatinine Units Must Match the Equation

The bedside Schwartz coefficient 0.413 requires serum creatinine in mg/dL. If a laboratory reports creatinine in μmol/L, convert the measurement or use the algebraically equivalent coefficient for that unit. Since 1 mg/dL is approximately 88.4 μmol/L, substituting a μmol/L value directly into the mg/dL equation makes the denominator about 88.4 times too large and produces an erroneous estimate.

The equivalent equation follows directly from unit conversion. If $\mathrm{SCr}_{\mu mol/L}$ is the reported concentration, then $\mathrm{SCr}_{mg/dL}=\mathrm{SCr}_{\mu mol/L}/88.4$. Substitution into the bedside equation gives:

\frac{0.413 \times \mathrm{height}}{\mathrm{SCr}_{\mu mol/L}/88.4} = \frac{0.413 \times 88.4 \times \mathrm{height}}{\mathrm{SCr}_{\mu mol/L}} \approx \frac{36.51 \times \mathrm{height (cm)}}{\mathrm{SCr}_{\mu mol/L}}.

Worked Example: Avoiding a Creatinine Unit Error

A hypothetical child is 130 cm tall and has serum creatinine of 0.80 mg/dL, equivalent to 70.72 μmol/L. Calculate the eGFR correctly in either unit system, then compare with the result from an incorrect substitution.

Using mg/dL:

\mathrm{eGFR} = \frac{0.413 \times 130}{0.80} = \frac{53.69}{0.80} = 67.1\ \text{mL/min/1.73 m}^2.

Using μmol/L with the equivalent coefficient:

\mathrm{eGFR} = \frac{36.51 \times 130}{70.72} = 67.1\ \text{mL/min/1.73 m}^2.

If 70.72 were incorrectly substituted into the mg/dL version, the calculation would instead be $53.69/70.72=0.759$ mL/min/1.73 m², an implausible result caused by a unit mismatch. Checking the scale and units is an important diagnostic, but a plausible-looking value does not by itself establish that the chosen equation is appropriate.

From Estimate to Dose Category

After the estimate is calculated, assign a category only by following the rule that applies to the drug and patient. That rule may specify indexed eGFR, absolute eGFR, creatinine clearance, or another measure; it may also specify age limits, equation, assay, and how to handle a value at an exact boundary. Do not replace one measure with another merely because both are described as kidney function.

Suppose a hypothetical drug protocol states that for an indexed eGFR of 30 to less than 60 mL/min/1.73 m², the usual dose is reduced to 50%, while values of at least 60 receive the usual dose. If the appropriate indexed eGFR is 55 mL/min/1.73 m², that estimate falls in the reduced-dose category. The percentage reduction is part of this invented rule, not a general clinical recommendation. A dose calculation would then use the protocol's stated dose basis, schedule, and rounding instructions.

When the estimate lies close to a category boundary, confirm height, weight if needed, creatinine units, assay convention, equation, and the protocol's inequality signs. If clinical status or the estimate is inconsistent with the broader picture, seek an appropriate reassessment rather than treating the equation output as exact. A different validated estimate or measured GFR may be relevant when specified by the clinical question, but selecting among methods requires justification rather than simply choosing the result that gives the desired category.

Diagnostics, Alternatives, and Limitations

A reliable calculation has both numerical and clinical checks. The numerical checks are straightforward: the equation's required units match the entered data, the height is in centimeters, the correct coefficient is used, and the final unit is stated. For de-indexing, verify that BSA is in m² and that the conversion factor is BSA divided by 1.73, not 1.73 divided by BSA. Confirm whether the protocol uses the indexed or absolute value before comparing against its threshold.

The main alternatives are not interchangeable shortcuts. Another pediatric estimating equation may be more appropriate for a defined age range, population, marker, or laboratory method. A measured GFR may be chosen when a more direct assessment is needed and available. A dosing protocol may specify a particular renal-function metric for consistency with the evidence supporting its thresholds. The selection depends on intended use and validation; no single equation is universally best for every pediatric dose decision.

Estimated GFR has limitations because it is an estimate derived from a marker influenced by more than filtration. Low muscle mass can lower creatinine production and make creatinine-based eGFR appear higher; unusual muscle mass can affect it in the other direction. Non-steady-state creatinine, poor-quality height or laboratory data, and use outside an equation's validated population can also undermine interpretation. A result should therefore be reported with its equation and units, not as a bare number.

Common Mistakes

Wrong reasoning: “The reported eGFR is 82.6, so it exceeds a cutoff of 50.” Correction: First establish whether the cutoff is in mL/min or mL/min/1.73 m²; de-index only when an absolute value is required.
Wrong reasoning: “The older coefficient gives a higher estimate, so it is safer to use.” Correction: Use the equation specified and validated for the setting. Coefficients reflect equation development and assay conventions, not adjustable dose-safety margins.
Wrong reasoning: “Creatinine is a number, so it can be entered in any reported unit.” Correction: Convert to the units required by the equation or use its correctly converted equivalent.
Wrong reasoning: “A calculated eGFR is the child's exact GFR.” Correction: It is an estimate with assumptions and limitations; clinical circumstances may require reassessment or another specified method.

Key takeaway: A dose category is defensible only when the equation, assay convention, creatinine units, body-size scaling, and threshold all refer to compatible quantities. Calculate carefully, state the result with units, and apply the dosing rule exactly as specified.

Check Your Understanding

Answer each question, then compare your reasoning with the solutions.

A hypothetical child is 110 cm tall and has stable serum creatinine of 0.50 mg/dL. Calculate bedside Schwartz eGFR and state its units.
A protocol gives a cutoff in mL/min, but the available eGFR is indexed in mL/min/1.73 m². What additional information and calculation are needed before comparing with the cutoff?
A laboratory reports creatinine in μmol/L. Why is it incorrect to substitute that value directly into the bedside equation with coefficient 0.413?
Two equations give estimates on opposite sides of a hypothetical dose boundary. What should determine the equation used for dose categorization?

Solutions. (1) $\mathrm{eGFR}=(0.413 \times 110)/0.50=45.43/0.50=90.9$ mL/min/1.73 m². (2) Obtain the child's height and weight, calculate BSA using $\sqrt{(\text{height in cm}\times\text{weight in kg})/3600}$, then multiply indexed eGFR by BSA/1.73 to obtain mL/min. (3) The coefficient 0.413 requires mg/dL; use the conversion 1 mg/dL $\approx 88.4$ μmol/L or the equivalent coefficient 36.51 for μmol/L. (4) Use the equation and renal-function measure specified and validated by the applicable drug protocol; do not choose an equation solely because its result gives a preferred category.