For most adult men a normal serum creatinine is roughly 0.7 to 1.3 mg/dL, and for most adult women roughly 0.6 to 1.1 mg/dL. In the units used across the UK, Europe and much of the world that is about 62 to 115 µmol/L for men and 53 to 97 µmol/L for women. Those two lines answer the question, but they hide almost everything that makes the answer useful: how the ranges change with age, why women sit lower, what happens in pregnancy, why children need entirely different numbers, and why a “normal” result is a statement about a population rather than a verdict on you.
This page is the numbers page. Everything below is reference ranges: by sex, by age from the first day of life to the ninth decade, in both unit systems side by side, in pregnancy trimester by trimester, and in children. Then the part that matters more than any table, which is what the word “normal” actually means when a laboratory prints it next to your result, how those ranges were built in the first place, and why your own previous readings tell you more than any published interval ever will.
One boundary is worth drawing straight away. This article covers the creatinine level in your blood, measured in mg/dL or µmol/L. It does not cover creatinine clearance, which is a different measurement in different units, expressed in mL/min, with its own set of normal values covered at normal creatinine clearance range. If you want a plain explanation of the molecule itself before the numbers, what creatinine is covers that ground, and questions about whether a particular result is a problem belong to when to worry about creatinine levels.
A creatinine value on its own says little without your age, sex and weight beside it. The Waldev creatinine clearance calculator converts your result into an estimate of how much filtration it actually represents, which is the figure clinicians work from.
On this page
The standard adult reference ranges
Start with the figures most laboratories in the United States, the United Kingdom and Australia print on an adult report. They vary a little between sites, and the section on laboratory differences further down explains why, but these are representative.
| Group | mg/dL | µmol/L | Notes |
|---|---|---|---|
| Adult men | 0.7 – 1.3 | 62 – 115 | The most widely quoted adult male interval |
| Adult women | 0.6 – 1.1 | 53 – 97 | Lower across the whole range, chiefly muscle mass |
| Adolescent boys (13–18) | 0.5 – 1.0 | 44 – 88 | Rises steeply through puberty |
| Adolescent girls (13–18) | 0.5 – 0.8 | 44 – 71 | Rises far less; the sex gap opens here |
| Some UK laboratories, men | 0.7 – 1.2 | 59 – 104 | Enzymatic assays often report slightly tighter |
| Some UK laboratories, women | 0.5 – 1.0 | 45 – 84 | Same assay effect |
Notice that the male and female intervals overlap heavily. A value of 1.0 mg/dL sits comfortably inside both. What differs is where each sits within its own range: 1.0 is mid-range for a man and near the top for a woman, and that positional difference carries information that a simple in-or-out verdict throws away.
The other thing the table does not show is how narrow these ranges are in absolute terms. The entire adult female interval spans half a milligram per decilitre. Laboratory assays carry a few percent of analytical variation, and your own body varies by a similar amount from week to week, which means the difference between 1.0 and 1.1 is often no difference at all. That has real consequences for how much weight anyone should put on a single decimal place, and it comes up again in the section on borderline results.
mg/dL and µmol/L: the same number in two costumes
A great deal of confusion in online searching comes from people comparing results measured in different units without realising it. Someone reads that 106 is high, panics about their own 1.06, and the two figures are the same value. The United States, and a handful of other countries, report creatinine in milligrams per decilitre. Most of the rest of the world, including the UK, Ireland, Canada, Australia and continental Europe, reports in micromoles per litre.
µmol/L = mg/dL × 88.4
mg/dL = µmol/L ÷ 88.4
The 88.4 comes from creatinine’s molecular weight of 113.12 grams per mole and the decilitre-to-litre conversion. It is exact enough for any practical purpose. Rounding to 88 changes nothing that matters clinically. Here is the conversion across the range you are likely to encounter.
| mg/dL | µmol/L | Typical interpretation in an adult |
|---|---|---|
| 0.4 | 35 | Below most adult ranges |
| 0.5 | 44 | Low-normal for a woman, low for a man |
| 0.6 | 53 | Bottom of the female range |
| 0.7 | 62 | Bottom of the male range, mid-low female |
| 0.8 | 71 | Comfortably mid-range for both |
| 0.9 | 80 | Mid-range male, upper-mid female |
| 1.0 | 88 | Mid male, near the female ceiling |
| 1.1 | 97 | Upper male, top of the female range |
| 1.2 | 106 | Upper male, above most female ranges |
| 1.3 | 115 | Top of the male range |
| 1.5 | 133 | Above range in both sexes |
| 2.0 | 177 | Clearly raised |
| 3.0 | 265 | Substantially raised |
A shortcut worth memorising if you switch between systems often: dividing a µmol/L figure by 90 and accepting a small error gets you close enough to the mg/dL equivalent for a quick sanity check. 120 µmol/L becomes about 1.33 mg/dL. The true answer is 1.36. Nothing in clinical decision-making turns on that gap.
Ranges by age, from birth to over 75
Creatinine tracks muscle mass, and muscle mass follows a life-long arc: almost none at birth, a steep climb through childhood and adolescence, a plateau across early and middle adult life, then a slow decline from roughly the fifth decade onwards. The blood level follows that curve, which is why a single adult range applied across all ages is a blunt instrument.
| Age band | mg/dL (approximate) | µmol/L (approximate) | What drives it |
|---|---|---|---|
| Day 1–3 (newborn) | 0.3 – 1.0 | 27 – 88 | Largely reflects the mother’s level, not the baby’s kidneys |
| Week 1–4 | 0.2 – 0.6 | 18 – 53 | Falls rapidly as the newborn’s own filtration establishes |
| 1–12 months | 0.2 – 0.4 | 18 – 35 | Tiny muscle mass, immature but functioning kidneys |
| 1–3 years | 0.2 – 0.5 | 18 – 44 | Muscle beginning to accumulate |
| 4–6 years | 0.3 – 0.6 | 27 – 53 | Steady climb with growth |
| 7–10 years | 0.4 – 0.7 | 35 – 62 | Still no meaningful sex difference |
| 11–13 years | 0.5 – 0.9 | 44 – 80 | Puberty begins to separate boys and girls |
| 14–18 years, male | 0.6 – 1.1 | 53 – 97 | Rapid muscle gain; approaches adult male values |
| 14–18 years, female | 0.5 – 0.9 | 44 – 80 | Reaches adult female values earlier |
| 19–39 years, male | 0.7 – 1.3 | 62 – 115 | Peak muscle mass for most people |
| 19–39 years, female | 0.6 – 1.1 | 53 – 97 | Peak muscle mass |
| 40–59 years, male | 0.7 – 1.3 | 62 – 115 | Printed range unchanged; true values drift down slightly |
| 40–59 years, female | 0.6 – 1.1 | 53 – 97 | As above |
| 60–74 years, male | 0.7 – 1.3 | 62 – 115 | Muscle loss begins offsetting filtration loss |
| 60–74 years, female | 0.6 – 1.1 | 53 – 97 | As above |
| 75+ years, either sex | 0.6 – 1.2 | 53 – 106 | A “normal” value can conceal substantially reduced filtration |
These paediatric and age-banded figures are approximate and compiled from typical published intervals. They differ between laboratories, between assay methods and between reference populations, sometimes considerably in the youngest age groups. Use them to understand the shape of the curve, not to judge an individual result. The range printed on your own report is the one that applies to you.
What stands out is the flatness of the adult rows. Most laboratories print one interval for adult men and one for adult women, unchanged from 19 to 95. That is a deliberate simplification, and an unfortunate one, because it is precisely in the oldest patients that the simplification misleads most. A gradual fall in muscle mass and a gradual fall in filtration can cancel each other out on paper. The number stays put; the kidney does not.
Normal creatinine levels for women, and why they sit lower
The female reference interval is lower at both ends, typically 0.6 to 1.1 mg/dL against 0.7 to 1.3 for men. The gap is around 0.1 to 0.2 mg/dL, or roughly 9 to 18 µmol/L, which sounds trivial and is not.
The reason is muscle. Around 95 per cent of the body’s creatine pool sits in skeletal muscle, and creatinine is produced at a fairly constant daily fraction of that pool. Women on average carry less skeletal muscle than men of the same height and weight, so they produce less creatinine per day and run a lower steady-state blood level with identical kidney function. Nothing about the female kidney filters differently. The difference is entirely upstream, in how much waste is arriving.
Where the sex difference comes from
Lower average skeletal muscle mass means lower daily creatinine production. Two people with identical filtration will show different blood levels if one has more muscle.
Where it does not come from
Not hormones acting on the kidney, not a smaller kidney, not a different filtration rate. Female GFR per unit body surface area is broadly comparable to male.
This has a practical edge that catches people out. A woman with a result of 1.15 mg/dL will see a “high” flag on her report; a man with 1.15 will not. Same number, same physiology, different verdict, purely because the comparison group differs. Conversely, a very muscular woman, a competitive rower or powerlifter, may sit at 1.2 mg/dL permanently and get flagged at every blood test she ever has, with entirely normal kidneys.
The mirror-image problem is more common and less recognised. A small, sedentary woman in her seventies with little muscle might have a creatinine of 0.9 mg/dL, well within the printed range, and a genuinely reduced filtration rate. Her result never triggers a flag. This is one of several reasons that eGFR, which builds sex and age into its calculation, is now reported automatically alongside creatinine in most health systems. For how those estimating equations differ from one another, Cockcroft-Gault versus MDRD compares them directly.
If your own result has come back below the female range, the causes and significance are covered separately in what low creatinine means, which handles low muscle mass, pregnancy, liver disease and the other common explanations.
Pregnancy: the ranges shift, and the shift is downwards
Pregnancy changes kidney physiology substantially and early. Plasma volume expands by around 40 to 50 per cent by the middle of the second trimester. Renal blood flow increases. Glomerular filtration rate rises by roughly 40 to 50 per cent above the pre-pregnancy value, and much of that increase is established within the first trimester, well before the bump shows. More filtration plus more dilution means creatinine falls.
| Stage | Typical mg/dL | Typical µmol/L | Comment |
|---|---|---|---|
| Pre-pregnancy / non-pregnant | 0.6 – 1.1 | 53 – 97 | Standard adult female range |
| First trimester | 0.4 – 0.7 | 35 – 62 | GFR already substantially increased |
| Second trimester | 0.4 – 0.6 | 35 – 53 | Usually the lowest values of the pregnancy |
| Third trimester | 0.4 – 0.7 | 35 – 62 | Drifts up slightly towards term |
| Postpartum (weeks) | Returning to 0.6 – 1.1 | 53 – 97 | Normalises over several weeks |
The clinically important consequence is that a value which looks perfectly normal on a standard adult report can be abnormal in pregnancy. A creatinine of 1.0 mg/dL in a woman at 28 weeks is inside the printed range and is, in context, notably raised for where she should be. Obstetric guidance recognises this: a serum creatinine above 1.1 mg/dL (about 97 µmol/L), or a doubling of the baseline value in the absence of other kidney disease, is one of the criteria used in assessing pre-eclampsia with severe features.
Because pregnancy ranges are lower, a rising creatinine during pregnancy should always be interpreted against earlier pregnancy values rather than the general adult range. If you are pregnant and your creatinine has risen, even while still inside the standard interval, raise it with your midwife or obstetrician rather than reading it as reassurance.
A second point about pregnancy testing: the estimating equations used to produce eGFR were not developed or validated in pregnant populations, and eGFR reported during pregnancy is unreliable. Clinicians generally fall back on the raw creatinine, on trends within the pregnancy, and where necessary on measured collections. This is one of the clearest examples of a general truth in this whole area, which is that population formulas fail at the edges of the population they were built from.
Paediatric ranges, and why adult numbers must never be used
Applying an adult reference range to a child is one of the more consequential errors possible with this test, and it runs in a direction that hides problems rather than creating false alarms.
Consider a four-year-old with a creatinine of 0.6 mg/dL. Against the adult male range of 0.7 to 1.3, that reads as low, or at worst unremarkable. Against the range appropriate to a four-year-old, roughly 0.3 to 0.6, it sits at the ceiling and may represent significantly reduced filtration. A busy clinician glancing at an adult-referenced report would see nothing wrong. The child could be well into kidney impairment.
Newborns are not measuring their own kidneys. In the first two to three days of life, a baby’s creatinine largely reflects what crossed the placenta from the mother. It falls over the first week as the infant’s own filtration takes over. A single high value on day one is usually meaningless in isolation; the trajectory over the first week is what matters.
Preterm infants differ again. Babies born prematurely have immature tubular function and can show a transient rise in creatinine after birth before it falls, a pattern that is normal for them and would be alarming in a term baby.
Height matters more than weight in children. The paediatric estimating equations, particularly the Schwartz formula, use height rather than weight because it correlates better with muscle mass in a growing body. A short child and a tall child of the same age have genuinely different expected creatinine values.
The sex difference does not exist before puberty. Boys and girls track together until testosterone starts building male muscle mass in adolescence. Splitting paediatric ranges by sex below about eleven years adds nothing.
The rise through adolescence is fast. A boy’s creatinine can climb from around 0.6 to around 1.0 mg/dL over three or four years of pubertal growth, with no change in kidney function whatsoever. Interpreting a rise in a teenager without accounting for growth invites unnecessary investigation.
Any laboratory reporting on children should be using age-specific paediatric intervals, and most do. If you are looking at a child’s result and the report shows an adult range, that is worth querying with the requesting clinician before drawing any conclusion at all.
Older adults: when a normal number stops being reassuring
Take an 81-year-old woman, five feet one, frail, forty-five kilograms, on five regular medications. Her creatinine comes back at 1.0 mg/dL. The report shows no flag. The range says 0.6 to 1.1. She is, on paper, normal.
Run the same value through an estimating equation with her age and sex included and the picture changes: her filtration is likely to be somewhere around 50 mL/min/1.73m², which is stage 3a chronic kidney disease. The creatinine did not lie. It simply reflected the fact that she has very little muscle producing very little creatinine, so a substantially impaired kidney still keeps her blood level looking ordinary.
What the raw number shows
1.0 mg/dL, inside the printed adult female range, no flag on the report, easy to skim past on a busy clinic list.
What the context shows
Age 81, low body weight, minimal muscle. Estimated filtration around half of normal, with real implications for drug dosing and future monitoring.
This is not a rare edge case. It is the routine situation in geriatric medicine, and it is the single strongest argument for never reading a creatinine value in isolation. Where it bites hardest is medication safety, because a large number of drugs are cleared by the kidney and dosed according to estimated function rather than the raw creatinine. Missing a halved filtration rate in an older patient can mean months of accumulating drug at a dose their kidneys cannot handle. Creatinine clearance in drug dosing covers that in detail, and running the numbers through the CrCl calculator takes seconds.
The other direction exists too, and produces a great deal of unnecessary anxiety. A 32-year-old weightlifter, ninety-five kilograms of mostly muscle, taking creatine, may sit at 1.4 mg/dL year after year with flawless kidneys. His report flags him every time. Whether that pattern needs any action is the subject of what high creatinine means, and the full list of explanations sits in what causes high creatinine levels.
How a reference range is actually built
Almost nobody is told this, and it changes how the word “normal” should land.
A laboratory establishing a reference interval recruits a group of people it considers healthy. Standards bodies recommend a minimum of around 120 such individuals per group, and in practice larger sets are used where possible. Blood is taken, creatinine is measured on that laboratory’s specific analyser using its specific method, and the results are plotted as a distribution.
People judged healthy by defined criteria: no known kidney disease, no relevant medication, often excluding pregnancy and extremes of age. The definition of “healthy” is itself a judgement call, and different laboratories make it differently.
The interval is method-specific. A range derived on an enzymatic analyser does not transfer cleanly to a laboratory running the older Jaffe method, which is why ranges travel poorly between sites.
The interval is defined as the values between the 2.5th and 97.5th percentiles of that healthy distribution. Everything below the 2.5th percentile and above the 97.5th is excluded from the printed range.
Those two percentile values become the numbers printed on every report, and they are what the words “normal range” refer to. They were never a statement about health. They are a statement about where most healthy people happened to fall.
Sit with the third step, because everything else follows from it. If the interval is defined as the central 95 per cent of healthy people, then by construction 5 per cent of perfectly healthy people fall outside it. One in forty above the top, one in forty below the bottom. Not because anything is wrong with them, but because a distribution has tails and somebody has to be in them.
Reference interval = 2.5th percentile to 97.5th percentile of a healthy reference population
Therefore: 5% of healthy people fall outside their own reference interval, by definition
Extend that across a full blood panel and the arithmetic becomes uncomfortable. If a metabolic panel reports twenty separate analytes and each has an independent 5 per cent chance of falling outside its interval in a healthy person, the probability of at least one flagged result approaches 64 per cent. Most people who have a broad panel done will see at least one abnormal flag, and most of the time it means nothing. That is not a failure of laboratory medicine. It is what happens when you draw a line at the 97.5th percentile and then measure twenty things at once.
Why “normal” is a population statistic, not a personal target
The word does a lot of unearned work. On a report, “normal” means one thing only: your value fell between two percentile boundaries derived from a group of other people. It does not mean your kidneys are healthy, that nothing has changed, or that you have achieved something.
Two people illustrate the gap. The first has had creatinine of 0.7 mg/dL at every test for a decade and this year returns 1.1. Both figures are inside the female reference range. Every report said normal. Her value has risen by more than 50 per cent, which in filtration terms is a substantial loss, and the reference range was structurally incapable of telling her. The second has always measured 1.35, gets flagged every single time, and has never had a day of kidney trouble in his life.
The reference range answered a question neither of them asked. They wanted to know whether their kidneys were behaving as they had been. The range only knows whether they resemble a sample of strangers.
A useful reframe: think of the reference interval as a map of where most healthy people live, not a fence marking where you are allowed to be. Being outside the map is a prompt to look more closely. Being inside it is not proof that nothing has moved.
There is a second reason the population framing matters, which is that the reference population may not resemble you. Reference intervals are usually derived from adults recruited locally, and factors that shift creatinine, average muscle mass, body size, diet, the proportion of the sample who lift weights or eat large amounts of meat, differ between populations. A range derived in one setting and applied in another can be a poor fit. This is well recognised, and it is part of why the 2021 revision of the CKD-EPI equation removed the race coefficient that had previously been applied to eGFR calculations, on the grounds that race is a social category rather than a biological determinant of creatinine production.
Your own baseline beats any published range
If you take one thing from this page, take this. The most informative comparison for your creatinine is your own previous creatinine.
Creatinine is unusually well suited to this. In a stable person, the value is remarkably consistent over time, because it depends chiefly on your muscle mass and your filtration rate, and neither changes quickly. Studies of biological variation put the within-person coefficient of variation for serum creatinine at roughly 4 to 6 per cent, with analytical variation adding another 2 to 3 per cent. Combine them and you get a rough working rule: a change of less than about 10 to 15 per cent between two measurements is difficult to distinguish from noise. A change larger than that is likely to be real.
| Your usual value | Change likely to be noise | Change likely to be real | Inside the reference range? |
|---|---|---|---|
| 0.7 mg/dL | 0.65 – 0.78 | Above about 0.82 | Yes, until well past 1.1 |
| 0.9 mg/dL | 0.82 – 1.0 | Above about 1.05 | Yes, until past 1.1 or 1.3 |
| 1.1 mg/dL | 1.0 – 1.22 | Above about 1.27 | Borderline for a woman |
| 1.3 mg/dL | 1.18 – 1.44 | Above about 1.5 | Already at or past the ceiling |
Read the last column. In the first two rows, a change large enough to be genuinely meaningful still leaves the person inside the “normal” range, and no flag ever appears. The reference interval is silent exactly where a personal baseline speaks loudest.
Practically, this means the single most useful thing you can do with creatinine results is keep them. Note the value, the date, the units and the laboratory. Four or five results over a few years turn an isolated number into a line, and a line is interpretable in a way a point never is. If your surgery or health system gives you access to your record, the historical values are usually already there.
It also means being deliberate about the conditions under which you are tested. Heavy exercise in the preceding 48 hours, a large meat-based meal the night before, dehydration and creatine supplements all move the number without touching your kidneys. Testing under consistent conditions makes your own series comparable, which is the whole point of having one.
Why two laboratories print different ranges
Move house, change hospital, or have blood taken privately, and the range on the report may differ from the one you are used to. Three things drive that.
1. The assay method
The older Jaffe reaction measures creatinine by a colour change with alkaline picrate, and other substances in blood react too. It tends to read slightly higher than the enzymatic method, historically by around 0.1 to 0.2 mg/dL depending on the sample. Laboratories set their ranges to match their own method.
2. The reference population
Each laboratory derives its interval from people recruited in its own catchment, or adopts a published one. Different populations, different boundaries.
3. Rounding and local policy
Some sites report to one decimal place, some to two. Some publish a single adult range rather than separate male and female intervals. Some round the boundaries for readability.
The picture is better than it used to be. Since the mid-2000s, creatinine assays have been standardised against a reference method known as isotope dilution mass spectrometry, which pulled results from different manufacturers much closer together. That standardisation is why modern eGFR equations can be applied across laboratories at all. It reduced the spread; it did not abolish it.
What this means in practice is straightforward. Comparing two results from the same laboratory is reliable. Comparing results from different laboratories requires a little caution, particularly for small differences, and comparing a value against a range printed by a different laboratory is a mistake worth avoiding. Always read your result against the interval on the same page. For a plain-language overview of the test itself, the MedlinePlus creatinine test page is a reliable starting point.
What would a “good” or “healthy” creatinine level even be?
People search for a good creatinine level and a healthy creatinine level constantly, and the question contains a hidden assumption worth dismantling.
Creatinine is a waste product. It is the inert breakdown product of creatine phosphate in muscle, produced at a steady rate, doing nothing useful in the blood, cleared by the kidney. It is not a nutrient, not a hormone, not something the body regulates towards an optimum. There is no level at which it confers benefit and no level you should be aiming for. Unlike vitamin D or haemoglobin, where more or less genuinely matters to how you function, creatinine is simply a signal being read.
Lower is not automatically better. A very low creatinine usually reflects low muscle mass, and in an older or unwell person that carries its own poor prognosis. Frailty, malnutrition, advanced liver disease and prolonged immobility all drive creatinine down. Nobody should be pleased about a value of 0.4 mg/dL in a seventy-year-old.
Higher is not automatically worse. Muscular people run higher values with entirely normal kidneys. A rugby forward at 1.3 mg/dL is not less healthy than a sedentary office worker at 0.8.
The useful target is stability, not a number. A value that stays where it has always been, in someone whose muscle mass has not changed, is the reassuring pattern. Movement is the signal.
What you can actually influence is filtration, not creatinine. Blood pressure control, glucose control, sensible use of anti-inflammatory painkillers, adequate hydration and not smoking protect kidney function. Trying to push a number down for its own sake is aiming at the dial rather than the engine.
If a phrase must be attached, the honest version of “good creatinine level” is: a value consistent with your own muscle mass, stable across time, and corresponding to an adequate estimated filtration rate for your age. That is three conditions, none of which is a single number, which is why the question never has the tidy answer people are hoping for.
Borderline results: 1.1, 1.2, 1.35 and the space between
The most common real situation is not a dramatically abnormal result. It is a value a whisker outside the line, or a whisker inside it, and a person wondering what to do.
The first thing to absorb is that the boundary is not a cliff. Nothing physiological happens between 1.29 and 1.31 mg/dL. The line was drawn at the 97.5th percentile of a sample of healthy volunteers, and it could reasonably have been drawn a little either side. Treating it as a threshold with meaning of its own is treating a statistical convention as biology.
| Situation | What it usually means | Reasonable response |
|---|---|---|
| Just above the range, first time, no symptoms | Often hydration, diet, exercise or assay variation | Repeat after a few days of normal hydration and no heavy training |
| Just above the range, stable across several years | Frequently your personal normal, especially if muscular | Note it as your baseline; no repeated investigation needed unless it moves |
| Inside the range but well above your own usual value | A real change that the flag system cannot see | Mention the trend explicitly to your doctor, with previous values |
| Just below the range | Low muscle mass, pregnancy, sometimes liver disease | Context matters more than the number; see the low creatinine article |
| Above the range with new symptoms | Potentially an acute change | Contact a doctor promptly rather than waiting for a repeat |
The second decimal place deserves particular scepticism. Values like 1.48 and 1.42 are not meaningfully different from each other; the difference sits inside analytical variation. People spend a lot of energy on that digit and it carries almost no information.
What genuinely changes the interpretation of a borderline result is everything around it: whether protein or blood is present in the urine, what the potassium and bicarbonate are doing, what the blood pressure is, what medications were started recently, and whether there are symptoms. A borderline creatinine with a clean urine test and a normal blood pressure is a very different object from the same number with protein in the urine. The NIDDK guide to kidney disease testing sets out how those pieces fit together.
Normal creatinine level versus normal creatinine clearance
These get conflated constantly in search, and they are genuinely different measurements with different units and different normal values.
| Serum creatinine level | Creatinine clearance | |
|---|---|---|
| What it measures | Concentration of creatinine in blood | Volume of blood cleared of creatinine per minute |
| Units | mg/dL or µmol/L | mL/min |
| Typical adult normal | 0.6 – 1.3 depending on sex | Roughly 90 – 140 mL/min in men, 80 – 130 in women |
| Direction of abnormality | Higher usually means worse filtration | Lower usually means worse filtration |
| How obtained | Single blood sample | Estimated from a formula, or measured with a 24-hour urine collection plus blood |
| Main use | Screening, monitoring trends | Estimating filtration, drug dosing decisions |
The direction reversal trips people up more than anything else. A rising creatinine level and a falling clearance describe the same deterioration. If you have seen both figures on a report and they seem to contradict each other, that is why.
Clearance is the more clinically useful of the two, because it accounts for the age, sex and body size that make a raw creatinine ambiguous. Its normal values by age are set out at normal creatinine clearance range, the concept itself at what creatinine clearance is, and how clearance relates to the eGFR figure most reports now show at creatinine clearance versus GFR. To convert your own creatinine into a clearance estimate, the Waldev CrCl calculator does the arithmetic.
Reading the creatinine line on your own report
A typical report gives you four things on one line, and each carries a different weight.
Check the units before anything else. A number near 100 is almost certainly µmol/L; a number near 1 is mg/dL. Comparing across unit systems without converting is the most common self-inflicted scare in this whole topic.
This is the range that applies to your sample, on that analyser. Use it in preference to anything you read online, including the tables on this page.
H, L, an asterisk or a colour. It only tells you which side of a percentile boundary you fell on. It knows nothing about your muscle mass, your history or your previous results.
Most systems now calculate this automatically from your creatinine, age and sex. It is usually the more informative figure, particularly at the extremes of age and body size, and it is where a quiet decline shows up first.
Two additions make the line far more useful. First, look for your previous results and put them in order. Second, check whether a urine test was done at the same time, because protein in the urine changes the interpretation of any creatinine value substantially and is often the earlier signal of the two. The National Kidney Foundation’s eGFR explainer covers how that figure maps onto the recognised stages of kidney disease.
Symptoms sit above all of this. A creatinine result, however reassuring, does not override a marked reduction in urine output, new swelling of the legs or face, breathlessness, confusion, or persistent vomiting. Those need same-day medical assessment regardless of what any number said last week.
Frequently asked questions
What is a normal creatinine level?
For most adult men, roughly 0.7 to 1.3 mg/dL, which is about 62 to 115 µmol/L. For most adult women, roughly 0.6 to 1.1 mg/dL, or about 53 to 97 µmol/L. Laboratories differ by a small margin depending on the assay they run and the population their range was derived from, so the interval printed on your own report takes precedence over any figure quoted online. Children have entirely separate age-based ranges that are much lower, and pregnancy shifts the female range downwards by roughly a third.
What is the normal creatinine level for women?
Typically 0.6 to 1.1 mg/dL, or 53 to 97 µmol/L, though some laboratories publish 0.5 to 1.0. The female range sits lower than the male range at both ends because women on average carry less skeletal muscle, and muscle mass determines how much creatinine is produced each day. Female kidneys do not filter differently. During pregnancy the expected range falls further, to roughly 0.4 to 0.7 mg/dL, so a value that looks normal on a standard adult report can be raised for a pregnant woman.
What should my creatinine level be?
There is no target to aim for, because creatinine is an inert waste product rather than something your body regulates towards an optimum. The useful answer is that it should be consistent with your own muscle mass and stable over time. A muscular person sitting steadily at 1.3 mg/dL is in a better position than someone who has drifted from 0.7 to 1.1 while staying inside the printed range the whole way. Stability against your own baseline is the meaningful measure, not a specific figure.
Is a creatinine level of 1.0 normal?
Yes for both sexes, though it sits differently in each. For an adult man, 1.0 mg/dL is mid-range and unremarkable. For an adult woman it is inside the range but close to the upper boundary, which is worth noting if her previous results were 0.7 or 0.8. In a child, 1.0 would be clearly raised and needs paediatric ranges applied. In pregnancy, 1.0 is well above the expected level and should be discussed with your maternity team rather than read as normal.
Why do reference ranges differ between laboratories?
Three reasons. Laboratories use different assay methods, and the older Jaffe reaction tends to read slightly higher than the enzymatic method because other blood substances interfere with it. Each laboratory derives its interval from its own reference population, and populations differ in average muscle mass and body size. Local policy on rounding and decimal places adds a little more variation. Standardisation against isotope dilution mass spectrometry has narrowed the gaps considerably since the mid-2000s, but it has not removed them entirely.
Can a healthy person have a creatinine outside the normal range?
Yes, and it happens by design. A reference interval is defined as the central 95 per cent of a healthy reference population, meaning the values between the 2.5th and 97.5th percentiles. That leaves one in twenty healthy people outside their own range, half above and half below, purely because distributions have tails. Across a twenty-item blood panel, the chance of at least one flagged result in a completely healthy person approaches 64 per cent. A single flag is a prompt to look, not a diagnosis.
What is a normal creatinine level for a child?
It depends heavily on age and is much lower than adult values. Broadly, infants under a year sit around 0.2 to 0.4 mg/dL, children aged one to three around 0.2 to 0.5, four to six around 0.3 to 0.6, and seven to ten around 0.4 to 0.7. Adolescents climb towards adult values, with boys rising faster than girls through puberty. Applying an adult range to a child hides problems rather than creating false alarms, so paediatric results must always be read against age-specific intervals.
Does creatinine change with age?
The true value does, though most laboratories print a single adult range regardless. Levels climb steeply through childhood and adolescence as muscle accumulates, plateau through early and middle adulthood, then drift as muscle mass falls in later life. The complication in older adults is that declining muscle and declining filtration pull the number in opposite directions and can cancel out. A frail eighty-year-old can show a completely normal creatinine while having filtration around half of what it once was.
How much can creatinine vary between tests?
More than most people expect. Within-person biological variation runs at roughly 4 to 6 per cent, and analytical variation in the laboratory adds another 2 to 3 per cent. As a working rule, a change of less than about 10 to 15 per cent between two measurements is hard to distinguish from noise. Hydration, a large meat meal the night before, heavy exercise in the previous 48 hours and creatine supplements all move the number without any change in kidney function, which is why a single unexpected result is usually repeated.
Is a normal creatinine level a guarantee my kidneys are fine?
No, and this is the most important limitation of the test. Creatinine typically does not rise above the reference range until filtration has fallen substantially, so meaningful early kidney disease can sit entirely hidden behind a normal-looking number. It is particularly unreliable in people with low muscle mass, including many older adults. This is why eGFR is reported alongside it, why urine testing for protein matters so much, and why a rise within the normal range from your own baseline deserves attention.
The short version
Normal serum creatinine is roughly 0.7 to 1.3 mg/dL (62 to 115 µmol/L) for adult men and 0.6 to 1.1 mg/dL (53 to 97 µmol/L) for adult women, with much lower age-specific ranges in children, a downward shift in pregnancy to around 0.4 to 0.7 mg/dL, and a printed adult range that unhelpfully stays the same from nineteen to ninety-five. Women sit lower because they carry less muscle, not because their kidneys work differently.
The deeper point is what “normal” means. It is the central 95 per cent of a healthy sample, which guarantees that one in twenty healthy people falls outside it, and which says nothing about whether your own value has moved. Your previous results are more informative than any published interval. Keep them, compare against them, and use the CrCl calculator to translate a raw number into an estimate of filtration. More across the creatinine blog category, the wider health blog, the full health calculators library, and everything else at waldev.com.
Medical disclaimer: This article is general educational information about laboratory reference ranges and is not medical advice. The figures given are approximate, vary between laboratories and assay methods, and cannot be used to diagnose or exclude any condition in an individual. Never use this page to decide whether to seek care, delay care, or change any medication or treatment. Always interpret your results with the range printed on your own report and discuss them with a doctor or qualified healthcare professional. Seek urgent medical attention if you have much reduced urine output, new swelling, breathlessness, confusion or persistent vomiting.
MedlinePlus explains what the creatinine blood test measures, why it is ordered and what results mean. Creatinine test →
NIDDK on the blood and urine tests used to assess kidney function, and how they fit together. CKD tests & diagnosis →
The National Kidney Foundation on eGFR, its normal values and the five stages of kidney disease. Estimated GFR explained →