Paediatric Creatinine Reference Interval Interpreter

Q: Why can't an adult creatinine reference interval be used in a child?

Because it is far too high. The enzymatic dataset assessed by the IFCC gives 15 to 30 µmol/L for a 1-to-3-year-old and 35 to 53 µmol/L for a 9-to-11-year-old, against 64 to 104 µmol/L for an adult man. A 10-year-old with a creatinine of 60 µmol/L is above their own 97.5th centile and would be reported as LOW by a laboratory printing an adult range. The error reassures about the children who are not normal, which is the worst direction for it to run.

Q: Why is a neonate's creatinine higher than an infant's?

Because creatinine crosses the placenta freely, so a newborn's first values largely reflect the mother's, and because neonatal tubular handling reabsorbs creatinine while glomerular filtration is still immature. Cord serum runs 46 to 86 µmol/L and a term neonate in the first two weeks 22 to 73, before falling to 11 to 34 µmol/L between two months and a year — the lowest interval in the whole of life. A single neonatal creatinine therefore says little and the trend says a great deal.

Q: Can a creatinine inside the reference interval still mean reduced kidney function?

Yes, and this is the main reason the page exists. The relationship is reciprocal — bedside Schwartz gives eGFR as 0.413 × height in cm ÷ creatinine in mg/dL — so the first rise in creatinine costs the most filtration. For a 10-year-old of 140 cm, a creatinine of 0.60 mg/dL gives about 96 mL/min/1.73 m², 0.90 mg/dL gives about 64, and 1.10 mg/dL about 53. The last two would be reported as normal on any adult interval.

Q: Does sex change a child's creatinine reference interval?

Only from adolescence. Boys and girls increase equally up to about age 14, after which boys rise more steeply, and in the adult reference set age significantly affected creatinine while sex did not (P = 0.9523). One large series found male upper reference limits exceeding female ones only above 13 years of age, by 0.11 to 0.23 mg/dL. This page therefore sex-splits from 15 years and not below it.

Q: Does it matter whether the assay is enzymatic or Jaffe?

It matters most in exactly this low range. Jaffe methods showed interferent-related coefficients of variation of 8.0% to 27% at a creatinine of 40.4 µmol/L (0.46 mg/dL) against under 4% for enzymatic methods, with albumin, immunoglobulin G and fetal haemoglobin interfering with Jaffe and not with enzymatic. A direct comparison found Abbott Jaffe and enzymatic results differing by about 0.17 to 0.18 mg/dL in prepubertal children, which is most of a toddler's reference interval, and the published conclusion is that this can misclassify chronic kidney disease stage.

Q: Why is creatinine least reliable in the children you worry about most?

Because it is a product of muscle. Creatinine turnover increases with muscle mass, so a chronically ill, immobile, poorly grown or steroid-treated child generates less of it at any given filtration rate. The arithmetic then makes it worse rather than better: eGFR is height divided by creatinine, so a falsely low creatinine produces a falsely high eGFR. Cystatin C does not depend on muscle mass and is the better analyte where that is the clinical situation.

Serum creatinine read against age- and sex-specific reference intervals from an enzymatic, IDMS-anchored dataset — because an adult interval is badly wrong in a child, and a creatinine inside the child’s own interval can still sit on a substantially reduced GFR.

Last updated 16 Sep 2026

Serum creatinine against the age-specific interval

Creatinine + age band + sex → interpretation

Serum creatinineEnter the value as reported, and choose the matching unit below. The intervals on this page are from enzymatic, isotope-dilution-mass-spectrometry-anchored methods. If your laboratory uses a Jaffe (alkaline picrate) method the numbers are not interchangeable at this end of the range: Jaffe against enzymatic differed by about 0.17 to 0.18 mg/dL in prepubertal children in one comparison, which is most of a toddler's entire reference interval.

UnitsValues entered in mg/dL are multiplied by 88.4017 to give µmol/L, which is the unit every interval on this page is stated in. A creatinine of 0.60 mg/dL is 53 µmol/L, and 1.00 mg/dL is 88 µmol/L.

Age bandThese are the bands the source dataset publishes, not a convenient partition. Note the deliberate gap: there is no band for 15 days to 2 months, because no enzymatic interval for that window was retrievable and interpolating between a term neonate's 22 to 73 µmol/L and an infant's 11 to 34 µmol/L would be inventing a number across an interval that is moving in the opposite direction from the one you would guess.

SexSex only changes the interval from 15 years upwards on this page, and that is a property of the sources rather than an omission. In the adult reference set, age significantly affected creatinine (P < 0.0001) while sex did not; and in the paediatric data, boys and girls increase equally up to about age 14, after which boys rise more steeply. Independently, one large paediatric series found upper reference limits in males exceeding those in females only above 13 years of age, by 0.11 to 0.23 mg/dL.

At or above the age-specific upper reference limitExample

Creatinine 60 µmol/L in a 10-year-old boy

Reference intervals used on this page

Every interval here was measured by an enzymatic method — Roche Creatinine Plus on a Hitachi 911 for the paediatric bands, the same reagent on a Hitachi 717 for the adults, and Roche CREP2 on a COBAS INTEGRA for the adolescent band. The method is named because it has to be: Jaffe and enzymatic creatinine are not interchangeable at these concentrations. Note the shape of the table — the interval FALLS from the neonate to the infant and then climbs steadily with muscle mass, so it is not monotonic and cannot be interpolated by eye.
Age band	2.5th to 97.5th centile (µmol/L)	Equivalent (mg/dL)	Source
Cord serum	46 – 86	0.52 – 0.97	Ceriotti/IFCC Table 3 (enzymatic, n = 51)
Preterm neonate, 0 to 21 days	29 – 90	0.33 – 1.01	Ceriotti/IFCC Table 3 (n = 58)
Term neonate, 0 to 14 days	22 – 73	0.26 – 0.81	Ceriotti/IFCC Table 3 (n = 69)
15 days to 2 months	no enzymatic interval retrievable	—	deliberately absent
2 months to under 1 year	11 – 34	0.15 – 0.37	Ceriotti/IFCC Table 3 (n = 41)
1 to under 3 years	15 – 30	0.18 – 0.33	Ceriotti/IFCC Table 3 (n = 45)
3 to under 5 years	21 – 34	0.24 – 0.39	Ceriotti/IFCC Table 3 (n = 41)
5 to under 7 years	26 – 40	0.29 – 0.46	Ceriotti/IFCC Table 3 (n = 43)
7 to under 9 years	31 – 46	0.34 – 0.53	Ceriotti/IFCC Table 3 (n = 46)
9 to under 11 years	35 – 53	0.39 – 0.60	Ceriotti/IFCC Table 3 (n = 47)
11 to under 13 years	38 – 59	0.43 – 0.67	Ceriotti/IFCC Table 3 (n = 42)
13 to under 15 years	41 – 65	0.47 – 0.73	Ceriotti/IFCC Table 3 (n = 38)
15 to under 18 years, male	44 – 88	0.50 – 1.00	Thai paediatric enzymatic study
15 to under 18 years, female	33 – 68	0.37 – 0.77	Thai paediatric enzymatic study
18 years and over, male	64 – 104	0.72 – 1.18	Ceriotti/IFCC Table 2 (n = 120)
18 years and over, female	49 – 90	0.55 – 1.02	Ceriotti/IFCC Table 2 (n = 120)

What an adult interval does to a child's result

The middle column is what a laboratory that prints one all-ages interval reports. Four of these six results would be mis-flagged, and the two that matter most — a 10-year-old at 60 and at 80 µmol/L — are reported as low-normal and normal while sitting above the child's own 97.5th centile.
Creatinine	Verdict on the adult male interval (64 – 104 µmol/L)	Verdict on the age-specific interval
25 µmol/L in a 2-year-old	Low	Inside 15 – 30 µmol/L — normal
60 µmol/L in a 2-year-old	Low-normal	Twice the upper limit of 30 µmol/L
60 µmol/L in a 10-year-old	Low-normal	Above the upper limit of 53 µmol/L
80 µmol/L in a 10-year-old	Normal	Half again above the upper limit of 53 µmol/L
45 µmol/L in a 16-year-old girl	Low	Inside 33 – 68 µmol/L — normal
80 µmol/L in a term neonate on day 1	Normal	Inside 22 – 73? No — above it, but probably maternal and falling

Why a normal creatinine can sit on a reduced GFR

This is the site's own arithmetic on the bedside Schwartz equation, eGFR = 0.413 × height in cm ÷ creatinine in mg/dL, and it is shown to make the direction visible rather than as calibrated estimates — bedside Schwartz was derived in children with chronic kidney disease and measured GFRs of roughly 15 to 75, so the upper rows are extrapolations. The point survives the caveat: every creatinine in the last three rows would be reported as normal by an adult interval, and every one of them implies a filtration rate a third to a half below where this child should be.
10-year-old, height 140 cm	Creatinine	Bedside Schwartz eGFR
Middle of the 9-to-11-year interval	0.50 mg/dL (44 µmol/L)	116 mL/min/1.73 m²
Top of the 9-to-11-year interval	0.60 mg/dL (53 µmol/L)	96 mL/min/1.73 m²
Adult female interval, lower half	0.70 mg/dL (62 µmol/L)	83 mL/min/1.73 m²
Adult male interval, centre	0.90 mg/dL (80 µmol/L)	64 mL/min/1.73 m²
Adult male interval, upper end	1.10 mg/dL (97 µmol/L)	53 mL/min/1.73 m²

Jaffe against enzymatic, in the range that matters

A method difference of 0.17 mg/dL on a 1-to-3-year-old's reference interval of 0.18 to 0.33 mg/dL is most of the interval. That is why every interval on this page names its method, and why the published conclusion is that inter-assay variability in children can lead to chronic kidney disease misclassification. Fetal haemoglobin interfering with Jaffe is a specifically neonatal reason to know which assay produced the number.
Comparison	Figure
Jaffe within-laboratory interferent-related CV at 40.4 µmol/L (0.46 mg/dL)	8.0% to 27%
Enzymatic within-laboratory interferent-related CV at the same concentration	Under 4%
Substances interfering with Jaffe and not with enzymatic	Albumin, immunoglobulin G, fetal haemoglobin
Abbott Jaffe against enzymatic, prepubertal females	0.18 mg/dL (2.5th to 97.5th centile −0.30 to −0.06)
Abbott Jaffe against enzymatic, prepubertal males	0.17 mg/dL (−0.30 to −0.04)
Abbott Jaffe against enzymatic, pubertal females and males	0.12 and 0.11 mg/dL
Six IDMS-traceable assays at a creatinine of 0.273 mg/dL	4 of 6 failed a minimal bias specification of 8%; biases −7.5% to 86%

A low interval, a reciprocal relationship, and a marker made of muscle

An adult serum creatinine reference interval is not merely imprecise in a child, it is the wrong interval by a wide margin. The enzymatic, isotope-dilution-mass-spectrometry-anchored dataset assessed by the IFCC gives a 1-to-3-year-old an interval of 15 to 30 micromoles per litre, against 64 to 104 for an adult man. A two-year-old with a creatinine of 60 has twice the upper limit for age and would be flagged LOW by a laboratory printing an adult range. So would a ten-year-old at 60, whose own upper limit is 53. The direction of the error is the dangerous one: the adult interval systematically reassures about children who are not normal.

The interval also does not move in the direction most people expect. It is high in the neonate — cord serum 46 to 86, preterm neonate 29 to 90, term neonate 22 to 73 micromoles per litre — because creatinine crosses the placenta freely and the first days' value is largely the mother's, and because neonatal tubular handling reabsorbs it. It then falls to 11 to 34 in the infant of two months to a year, which is the lowest interval in the whole of life, and climbs steadily from there as muscle accumulates. The curve is not monotonic, so it cannot be interpolated by eye, and this page leaves a deliberate hole between fifteen days and two months rather than guessing across the turning point.

Then there is the arithmetic of the relationship itself, which is why a creatinine inside a child's own interval is still not a statement about kidney function. The bedside Schwartz equation puts eGFR at 0.413 times height in centimetres divided by creatinine in milligrams per decilitre, so filtration and creatinine are reciprocal and the first increments of creatinine buy the largest losses of filtration. For a ten-year-old of 140 centimetres, a creatinine at the top of the age interval, 0.60 mg/dL, corresponds to about 96 mL/min/1.73 m². A creatinine of 0.90 — the middle of the adult male interval, normal on any adult report — corresponds to about 64. A creatinine of 1.10 corresponds to about 53. All three would be printed without a flag by a laboratory using one interval for everybody, and the last two describe a child who has lost a third to a half of their filtration.

Two further things limit the interval, and both are worth knowing before acting on a result. Creatinine is generated by muscle: turnover rises with muscle mass, so muscular people have higher levels and undernourished people lower ones, and a child's creatinine rises through childhood mainly because muscle is accumulating. That makes creatinine least reliable in exactly the children whose kidneys are most often in question — the chronically ill, the immobile, the poorly grown, the steroid-treated — and the failure is one-directional, because a falsely low creatinine divided into a height produces a falsely high eGFR. Cystatin C does not depend on muscle mass and is the better analyte when that is the situation. And the method matters more here than anywhere else in clinical chemistry: Jaffe methods showed interferent-related coefficients of variation of 8 to 27 per cent at a creatinine of 0.46 mg/dL against under 4 per cent for enzymatic ones, with albumin, immunoglobulin G and fetal haemoglobin among the interferents; and a direct comparison found Abbott Jaffe and enzymatic results differing by around 0.17 to 0.18 mg/dL in prepubertal children. On a toddler's interval of 0.18 to 0.33 mg/dL, that is most of the interval. Every interval on this page names its method for that reason.

Frequently asked questions

Why can't an adult creatinine reference interval be used in a child?

Because it is far too high. The enzymatic dataset assessed by the IFCC gives 15 to 30 µmol/L for a 1-to-3-year-old and 35 to 53 µmol/L for a 9-to-11-year-old, against 64 to 104 µmol/L for an adult man. A 10-year-old with a creatinine of 60 µmol/L is above their own 97.5th centile and would be reported as LOW by a laboratory printing an adult range. The error reassures about the children who are not normal, which is the worst direction for it to run.

Why is a neonate's creatinine higher than an infant's?

Because creatinine crosses the placenta freely, so a newborn's first values largely reflect the mother's, and because neonatal tubular handling reabsorbs creatinine while glomerular filtration is still immature. Cord serum runs 46 to 86 µmol/L and a term neonate in the first two weeks 22 to 73, before falling to 11 to 34 µmol/L between two months and a year — the lowest interval in the whole of life. A single neonatal creatinine therefore says little and the trend says a great deal.

Can a creatinine inside the reference interval still mean reduced kidney function?

Yes, and this is the main reason the page exists. The relationship is reciprocal — bedside Schwartz gives eGFR as 0.413 × height in cm ÷ creatinine in mg/dL — so the first rise in creatinine costs the most filtration. For a 10-year-old of 140 cm, a creatinine of 0.60 mg/dL gives about 96 mL/min/1.73 m², 0.90 mg/dL gives about 64, and 1.10 mg/dL about 53. The last two would be reported as normal on any adult interval.

Does sex change a child's creatinine reference interval?

Only from adolescence. Boys and girls increase equally up to about age 14, after which boys rise more steeply, and in the adult reference set age significantly affected creatinine while sex did not (P = 0.9523). One large series found male upper reference limits exceeding female ones only above 13 years of age, by 0.11 to 0.23 mg/dL. This page therefore sex-splits from 15 years and not below it.

Does it matter whether the assay is enzymatic or Jaffe?

It matters most in exactly this low range. Jaffe methods showed interferent-related coefficients of variation of 8.0% to 27% at a creatinine of 40.4 µmol/L (0.46 mg/dL) against under 4% for enzymatic methods, with albumin, immunoglobulin G and fetal haemoglobin interfering with Jaffe and not with enzymatic. A direct comparison found Abbott Jaffe and enzymatic results differing by about 0.17 to 0.18 mg/dL in prepubertal children, which is most of a toddler's reference interval, and the published conclusion is that this can misclassify chronic kidney disease stage.

Why is creatinine least reliable in the children you worry about most?

Because it is a product of muscle. Creatinine turnover increases with muscle mass, so a chronically ill, immobile, poorly grown or steroid-treated child generates less of it at any given filtration rate. The arithmetic then makes it worse rather than better: eGFR is height divided by creatinine, so a falsely low creatinine produces a falsely high eGFR. Cystatin C does not depend on muscle mass and is the better analyte where that is the clinical situation.

Related calculators

References

Ceriotti F, et al. Reference intervals for serum creatinine concentrations: assessment of available data for global application. Clin Chem. 2008;54(3):559 — IFCC Committee on Reference Intervals and Decision Limits; Table 3 (paediatric bands from cord serum to 13–15 years, enzymatic Creatinine Plus on a Hitachi 911, n = 521) and Table 2 (adults, 120 men and 120 women, IDMS-traceable).
Suwanrungroj S, Pattarapanitchai P, Chomean S, Kaset C. Establishing age and gender-specific serum creatinine reference ranges for Thai pediatric population. PLoS One. 2024;19:e0300369 — Roche CREP2 enzymatic colorimetric on a COBAS INTEGRA, 27,642 results; the sex-split 15-to-18-year band used here.
Establishing age/sex related serum creatinine reference intervals from hospital laboratory data based on different statistical methods. Clin Chim Acta. 2008 — IDMS-traceable enzymatic method; equal increases in boys and girls to age 14 and a steeper rise in boys from 14 to 20.
Serum creatinine reference limits in pediatric population — a single center electronic health record-based database in Taiwan. Front Pediatr. 2021;9:793446 — 27,911 subjects, compensated kinetic Jaffe; male upper reference limits exceeding female ones only above 13 years, by 0.11 to 0.23 mg/dL, and creatinine rising with muscle mass from 1 month of age.
Prime time for enzymatic creatinine methods in pediatrics. Clin Chem. 2009;55(3):549 — Jaffe interferent-related CVs of 8.0% to 27% at 40.4 µmol/L against under 4% for enzymatic methods, with albumin, IgG and fetal haemoglobin as interferents.
Large inter-assay difference of serum creatinine in pediatric population: a threat to accurate staging of chronic kidney disease. Pediatr Nephrol. 2022. doi:10.1007/s00467-021-05335-x — Abbott Jaffe against enzymatic differing by 0.11 to 0.18 mg/dL by pubertal status and sex, and 4 of 6 IDMS-traceable assays failing an 8% bias specification at 0.273 mg/dL.
Schwartz GJ, Muñoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629 — the bedside Schwartz equation and its 0.413 constant, derived in children with measured GFR of roughly 15 to 75 mL/min/1.73 m².
Merck Manual Professional Edition. Evaluation of the Renal Patient — creatinine turnover increasing with muscle mass, and serum creatinine alone as a poor indicator of kidney function.

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.