Creatinine on a blood test is a waste product your muscles make, measured in your blood to estimate how well your kidneys are filtering. It is one line on a panel, usually printed as “Creatinine, Serum” with a number, a unit, and a reference range beside it. This page is about the test itself: why it was ordered, which panels contain it, whether you need to fast, how the laboratory actually measures it, and how to read the result you were sent.
The creatinine blood test measures the concentration of creatinine in your bloodstream. Creatinine is produced at a fairly steady rate as muscle uses creatine for energy, and healthy kidneys filter it out into urine almost as fast as it appears. So the amount left circulating is a proxy for filtration. If the kidneys slow down, creatinine accumulates and the number rises. That single relationship is why this unremarkable-looking line appears on more blood panels than almost any other kidney measure.
Almost nobody has a creatinine test on its own. It arrives bundled into a basic metabolic panel, a comprehensive metabolic panel, a renal profile, or the set of tests a British laboratory calls urea and electrolytes. Your doctor may have ordered the panel for a reason that had nothing to do with your kidneys, and creatinine simply came along with it. If you want the biology rather than the procedure, what creatinine is covers the molecule itself, and what a high creatinine means deals with an out-of-range result.
Feed the number from your report into the Waldev creatinine clearance calculator to see the estimated clearance it corresponds to for your age, sex and weight. It accepts both mg/dL and µmol/L.
On this page
What the creatinine test actually measures
Strictly speaking, it measures concentration: how many milligrams of creatinine sit in each decilitre of your blood, or how many micromoles in each litre. That is all the machine reports. Everything else is inference.
The inference works because of a steady supply. Roughly 1 to 2 per cent of the body’s creatine pool converts spontaneously to creatinine every day, and the great majority of that pool lives in skeletal muscle. Production is therefore reasonably constant from day to day in the same person. Against that constant input, the blood level is set almost entirely by how quickly it leaves — and it leaves through the kidneys. Filtration falls, concentration climbs. The relationship is close enough to be clinically useful and loose enough to be repeatedly misread, which is why the rest of this cluster exists.
One thing the test does not measure is kidney damage. It measures kidney function, and function can be preserved for a long time while damage accumulates. A person can lose a substantial fraction of their filtering units and still return a creatinine inside the reference range, because the remaining units compensate by filtering harder. That is the well-known blind spot of this test and the reason a urine albumin measurement is usually ordered alongside it rather than instead of it.
A useful mental model: creatinine is a level in a tank with a steady tap running in and a drain letting it out. The test tells you the level. It does not directly tell you the size of the drain, and it certainly does not tell you why the drain changed.
Why your doctor ordered it
There is rarely a single dramatic reason. Creatinine sits on standard panels because it is cheap, fast, and informative about an organ that fails quietly. Here are the situations that actually generate the request.
| Reason for testing | What is being asked |
|---|---|
| Routine health check or annual review | Is baseline kidney function normal, and has it changed since last time? |
| Diabetes or high blood pressure monitoring | Are the two leading causes of chronic kidney disease starting to affect the kidneys? |
| Before starting a new medicine | Is filtration good enough for the usual dose, or does the dose need adjusting? |
| Ongoing monitoring on certain drugs | Diuretics, ACE inhibitors, ARBs, lithium, methotrexate and others all require periodic checks |
| Symptoms suggesting kidney trouble | Swelling, reduced urine output, unexplained fatigue, nausea, foamy urine |
| Before a contrast-enhanced CT or an MRI with gadolinium | Is filtration low enough to change how the scan is done? |
| Hospital admission or before surgery | A baseline, and a check for acute kidney injury developing during the stay |
| Known kidney disease | Is the trajectory stable or declining, and how fast? |
| Investigating an abnormal result elsewhere | Anaemia, high potassium, high calcium and unexplained weight loss all prompt a kidney check |
Notice how many of these are about medication rather than about kidney disease. Dose adjustment is one of the most common practical uses of the result, because a large number of drugs are cleared by the kidney and become dangerous if filtration is reduced and the dose is not. Creatinine clearance in drug dosing goes through how prescribers use the number for exactly that.
It is also worth saying plainly that being sent for this test is not a signal that anyone suspects kidney failure. In the overwhelming majority of cases it is included by default. If your doctor were specifically worried about your kidneys, you would generally be sent for a urine test at the same time, and probably an ultrasound.
Which panels include creatinine
Yes, a basic metabolic panel includes creatinine. So does everything larger. This is the question people ask most often after being handed a form, so here is the full picture.
| Panel | Where used | Contains creatinine? | What else is in it |
|---|---|---|---|
| Basic metabolic panel (BMP) | United States | Yes | Eight tests: glucose, calcium, sodium, potassium, chloride, bicarbonate, urea nitrogen (BUN), creatinine |
| Comprehensive metabolic panel (CMP) | United States | Yes | Fourteen tests: everything in the BMP plus albumin, total protein, and the liver tests ALP, ALT, AST and bilirubin |
| Renal function panel | United States | Yes | The BMP tests plus albumin and phosphate, aimed specifically at kidney assessment |
| Urea and electrolytes (U&E) | United Kingdom, Ireland, Australia | Yes | Sodium, potassium, urea, creatinine, usually with a calculated eGFR; sometimes chloride and bicarbonate |
| Liver function tests (LFTs) | Everywhere | No | Ordered separately; the CMP happens to bundle both |
| Full blood count / CBC | Everywhere | No | Different discipline entirely — cells rather than chemistry |
The practical consequence is that the same blood tube produces the creatinine result no matter which of these was requested. There is no separate needle, no separate appointment, and usually no extra cost, because the analyser runs the whole panel from one sample in a single pass.
British readers sometimes get confused by the terminology gap. A U&E is not a smaller test than a BMP in any meaningful sense; it just happens to omit glucose and calcium and to include a calculated eGFR by default, which American panels historically did not. Both are answering the same question about your kidneys.
Do you have to fast for a creatinine blood test?
No. Creatinine does not require fasting. You can eat and drink normally before it, and drinking water beforehand is actively helpful rather than something to avoid.
That answer needs one qualification, and it is the reason so many people are told to fast anyway. Creatinine almost never travels alone. If it is bundled into a comprehensive metabolic panel, the glucose measurement in that panel usually does want a fast of eight to twelve hours. If a lipid profile was ordered on the same form, older protocols asked for a fast there too, though many laboratories have now dropped that requirement. The fasting instruction attaches to the company creatinine keeps, not to creatinine itself.
So if you were told to fast, fast — the instruction exists for another test on the form. If you were told nothing, do not invent the rule. Arriving dehydrated after a twelve-hour fast with no water is more likely to distort a creatinine result than eating breakfast would be, because reduced circulating volume genuinely lowers filtration for a few hours.
Genuinely worth avoiding
A large portion of cooked meat in the twelve hours before the draw. Cooking converts creatine in muscle tissue into creatinine, which you absorb straight into your bloodstream. A big steak or a roast dinner can lift a morning result noticeably. Also worth avoiding: heavy resistance training or a long run in the previous 48 hours, and turning up dehydrated.
Genuinely fine
Water, in normal amounts. Black coffee or tea. A normal breakfast that is not meat-heavy. Your regular prescribed medicines, unless your doctor gave you specific instructions otherwise — never skip a prescription to “improve” a test. The time of day matters slightly but not enough to reschedule for.
There is one further nuance about supplements. If you take creatine monohydrate, you do not need to stop it, but you do need to tell whoever ordered the test. A larger creatine pool means more daily conversion to creatinine and a genuinely higher blood level, with kidney function completely unchanged. On a report it is indistinguishable from mild impairment. This is one of the more common sources of an unnecessary referral, and it is entirely avoidable with one sentence at the appointment.
What the test involves at the appointment
Very little. This is a standard venous blood draw and it takes under a minute once the needle is in.
You confirm your name and date of birth, which is how the tube gets labelled. Sample mislabelling is one of the few genuinely serious errors in laboratory medicine, so expect to be asked more than once.
Usually the antecubital fossa, the crease at the front of the elbow. The tourniquet stays on briefly; prolonged application can affect some analytes, so a good phlebotomist releases it early.
A few millilitres, typically into a gold or yellow-topped serum separator tube, or a green lithium heparin tube if the laboratory measures plasma. Multiple tubes are normal if several panels were requested.
A minute or two of pressure on the site prevents most bruising. Keep the arm straight rather than bending it up, which is the usual cause of the classic elbow-crease bruise.
The tube goes to the laboratory, is spun to separate cells from serum or plasma, and the clear fluid is loaded onto an automated chemistry analyser. Creatinine is stable enough that short transport delays do not matter much, unlike potassium.
A modern analyser runs hundreds of samples an hour. The creatinine result is generated, the eGFR calculated automatically from it, and the whole panel released to the requesting clinician.
The risks are the risks of any blood test and they are small: a bruise, brief soreness, occasionally feeling faint, very rarely a small collection of blood under the skin. If you know you faint at blood draws, say so before rather than after. Lying down for it is entirely routine and no one will think anything of it.
One laboratory detail that occasionally affects you: a haemolysed sample, where red cells have broken open in the tube, can interfere with certain measurements and may be rejected. If you are asked to come back and repeat a test with no explanation given, that is usually why. It reflects nothing about your health.
How the laboratory actually measures creatinine
This is the part almost nobody explains, and it turns out to matter for interpreting your number. There are two methods in routine use, and they do not always agree.
The Jaffe reaction
The older method, and still very widely used because it is cheap and fast. Creatinine reacts with alkaline picrate to form an orange-red complex, and the analyser measures the colour intensity. The chemistry dates to 1886, which tells you something about how durable a good assay can be.
Its weakness is specificity. Alkaline picrate reacts not only with creatinine but with a family of other substances in blood, collectively known as pseudo-creatinine chromogens: glucose, ketones, certain proteins, bilirubin, ascorbate, and some drugs including particular cephalosporin antibiotics. Every one of those can nudge the reported creatinine upwards without a single molecule of extra creatinine being present. Modern kinetic versions of the Jaffe method reduce the problem considerably by measuring the reaction rate over a specific time window rather than the endpoint, but they do not eliminate it.
Enzymatic assays
The more specific alternative. A cascade of enzymes — typically creatininase, creatinase and sarcosine oxidase — breaks creatinine down through a defined series of steps, generating hydrogen peroxide, which then drives a colour reaction that is measured. Because each enzyme only acts on its intended substrate, the interference from unrelated blood constituents largely disappears.
Enzymatic methods cost more per test, which is the only real reason they have not entirely displaced the Jaffe reaction. They are the preferred method in paediatrics, where values are low and a small absolute interference becomes a large proportional error, and in any situation where precision at the low end matters.
IDMS standardisation, and why it changed the numbers
Until around 2010, the same blood sample sent to different laboratories could come back with meaningfully different creatinine values, because each laboratory calibrated its assay independently. That is intolerable when a calculated eGFR depends on the number, so an international standardisation effort tied all routine assays back to a reference method: isotope dilution mass spectrometry, abbreviated IDMS.
IDMS is the gold standard. It separates creatinine by chromatography, ionises it, and quantifies it against a known amount of isotopically labelled creatinine added to the sample. It is far too slow and expensive for routine use, but it defines truth, and every routine analyser is now calibrated to agree with it.
The practical fallout was that standardised results run roughly 5 per cent lower than the older non-standardised Jaffe values. That is why the eGFR equations were recalibrated at the same time — the MDRD coefficient changing from 186 to 175 is the well-known example — and it is why comparing a creatinine from 2005 against one from today is not quite the like-for-like exercise it appears to be. If you are looking at old results, that is worth knowing. Cockcroft-Gault versus MDRD covers how the equations themselves differ.
Jaffe: creatinine + alkaline picrate → orange-red complex → measure absorbance
Enzymatic: creatinine → creatine → sarcosine → H₂O₂ → measured colour
Reference: IDMS, isotope dilution mass spectrometry — defines the true value
Finding creatinine on your report
Reports differ by laboratory, but the naming conventions are fairly consistent. Look for any of these.
Creatinine, Serum or Creatinine, Ser. The most common full form on American reports.
Creat or Cr. Abbreviated forms, common on printouts where column width is tight.
SCr. Serum creatinine, the form used most often in clinical writing and in the eGFR equations.
Creatinine (plasma). Some laboratories measure plasma rather than serum; the values are effectively interchangeable for this analyte.
Creatinine + eGFR. Increasingly the two are printed as a pair, since one is calculated from the other and neither is very useful alone.
Here is what the line typically looks like in context, with the neighbouring results that appear on the same block of a panel.
Two features of that mock report deserve comment. The reference range printed on the right is specific to that laboratory and that method, not a universal standard. And the eGFR immediately below is not a separate measurement at all; it is arithmetic performed on the creatinine value together with your age and sex. Nothing extra was measured to produce it.
mg/dL, µmol/L, and the number 88.4
Two unit systems are in use worldwide and they produce numbers that look nothing like each other. Conventional units, milligrams per decilitre, dominate in the United States. SI units, micromoles per litre, are used across the United Kingdom, Europe, Canada, Australia and most of the rest of the world.
µmol/L = mg/dL × 88.4
mg/dL = µmol/L ÷ 88.4
The factor is the molar mass of creatinine, 113.12 g/mol, run through the unit conversion. It is fixed, so the arithmetic is trivial once you know it. What it prevents is the specific panic of reading an American article, seeing “normal is around 1.0”, and then looking at your own British report that says 88 and assuming something has gone catastrophically wrong. Those are the same value.
| mg/dL | µmol/L | Rough interpretation in an average adult |
|---|---|---|
| 0.6 | 53 | Low end of normal |
| 0.8 | 71 | Comfortably normal |
| 1.0 | 88 | Normal for most adults |
| 1.2 | 106 | Upper end of normal for men; flagged high for many women |
| 1.5 | 133 | Above range in nearly all laboratories |
| 2.0 | 177 | Clearly reduced filtration |
| 3.0 | 265 | Substantially reduced filtration |
| 5.0 | 442 | Severe impairment |
That right-hand column is deliberately vague, and it should be. What a given value means depends heavily on your muscle mass, your age, and above all what your previous results were. When to worry about creatinine levels handles the threshold question properly. If you are converting a value to put into a calculator, note that the Waldev CrCl calculator accepts either unit and converts internally.
Reading the reference range and the H or L flag
The headline adult figures, which are approximate and vary by laboratory: roughly 0.7 to 1.3 mg/dL (about 62 to 115 µmol/L) for men, and roughly 0.6 to 1.1 mg/dL (about 53 to 97 µmol/L) for women. The difference is muscle mass, not any difference in kidney function between the sexes.
Ranges also shift with age and are much lower in children, which is a genuine source of alarm for parents comparing a child’s value against an adult scale. This article deliberately does not reproduce a full age-banded table — the normal creatinine clearance range guide owns that territory and does it properly.
What matters more here is understanding what the range actually is. A reference range is not a boundary between health and disease. It is a statistical interval, conventionally covering the central 95 per cent of results from a reference population that the laboratory considers healthy. Two consequences follow directly, and both are routinely forgotten.
Inside the range is not proof of health
Substantial loss of filtration can occur before creatinine crosses the upper limit, because surviving nephrons compensate. A move from 0.7 to 1.1 mg/dL is a large proportional change that generates no flag at all. This is why trend matters more than position.
Outside the range is not proof of disease
By construction, one healthy person in twenty falls outside a 95 per cent interval on any given test. Add a muscular build, a recent gym session or yesterday’s roast dinner and a mildly high flag becomes unsurprising. The single most common next step is simply to repeat it.
The H and L flags are generated automatically by comparing your number to the range stored for that assay. They carry no information about magnitude, urgency or cause. A creatinine 0.05 mg/dL above the limit and one three times the limit both get the same letter. Do not read the flag as a verdict; read the number, and read it next to your last one.
Many laboratories also apply a separate, much wider set of critical values that trigger a phone call to the requesting clinician regardless of the time of day. If nobody rang you, your result was not in that territory. That is a genuinely reassuring piece of information that patients rarely realise they have.
The tests printed next to it, and why they are there
Creatinine is interpreted as part of a group. Reading it in isolation discards most of the available information.
| Neighbouring result | What it adds |
|---|---|
| eGFR | Calculated from your creatinine, age and sex. Translates a concentration into an estimated filtration rate, which is what the CKD stages are actually based on. Not an independent measurement. |
| Urea (BUN in the US) | Another nitrogen waste product, but one strongly affected by hydration, protein intake and gut bleeding. A urea disproportionately high relative to creatinine points toward reduced blood flow rather than kidney damage. |
| Potassium | Rises when filtration falls badly. A high potassium alongside a high creatinine raises the urgency considerably, because it affects the heart. |
| Bicarbonate | Falls in the metabolic acidosis of advanced kidney disease. A quiet marker that the impairment is not trivial. |
| Sodium | Reflects fluid balance and helps distinguish dehydration from other causes. |
| Calcium and phosphate | Disturbed in chronic kidney disease as the kidney’s role in vitamin D activation and phosphate excretion is lost. |
| Albumin | Low albumin can accompany protein loss through damaged kidneys, and it also affects how calcium is interpreted. |
| Haemoglobin | Anaemia is common in chronic kidney disease because the failing kidney produces less erythropoietin. |
| Urine albumin-to-creatinine ratio | Not on the blood panel, but the essential companion test. Detects damage that blood creatinine misses entirely. |
That last row is the one to remember. Chronic kidney disease is staged on two axes, filtration and albuminuria, and a blood test alone only gives you one of them. If a raised creatinine was found and no urine test followed, asking about it is entirely reasonable. The NIDDK guidance on kidney disease testing sets out why the pair is used together.
The relationship between creatinine and estimated filtration is worth grasping properly, because it is not linear. A rise from 0.8 to 1.0 mg/dL barely moves the eGFR. A rise from 3.0 to 3.5 represents a much smaller proportional loss of remaining function than the earlier change did. Creatinine climbs slowly at first and then steeply, which flatters early disease and exaggerates late change. Creatinine clearance versus GFR and what creatinine clearance means both go further into this.
How long the result takes
Faster than most people expect at the bench, slower than most people want in practice, and the gap between those two facts causes a lot of unnecessary anxiety.
| Setting | Typical time to a result | What sets the delay |
|---|---|---|
| Hospital emergency department | Around 30 to 60 minutes | On-site analyser, samples run continuously, results push straight to the clinical system |
| Hospital inpatient, routine morning bloods | Usually by late morning | Batch processing of the morning round |
| Primary care with an on-site or local lab | Same day to 48 hours | Courier transport plus clinician review before release |
| Primary care sending to a regional lab | Two to five working days | Transport, batching, and the review step |
| Point-of-care device | About two minutes | Handheld cartridge analyser, whole blood, no centrifuge |
The analytical step itself is almost never the bottleneck. A chemistry analyser produces a creatinine in a couple of minutes. The delay is logistics and, importantly, clinician review: most systems hold results until a doctor has looked at them, so that an abnormal number reaches you with context rather than as a bare figure on a portal at eleven at night.
If you have access to a patient portal that releases results immediately, you may see the number before anyone has interpreted it. That is a mixed blessing. A single mildly high creatinine with no other change is a common and usually unremarkable finding, and the correct response is almost always to wait for the interpretation rather than to spend the evening reading about dialysis.
Point-of-care creatinine before contrast scans
A specific scenario worth explaining, because a lot of people meet creatinine testing for the first time in a radiology waiting room with a fingerstick device.
Iodinated contrast used in CT scans is cleared by the kidneys, and gadolinium agents used in MRI are too. Historically both were treated as significant kidney risks, and a recent creatinine was demanded before either. The evidence has moved. Large modern analyses suggest the risk of contrast-associated acute kidney injury from intravenous CT contrast was substantially overestimated, because earlier studies compared scanned patients against unscanned ones without accounting for how much sicker the scanned group was. Current consensus guidance concentrates concern on people with markedly reduced filtration, broadly an eGFR below 30, rather than on everyone.
For MRI, the specific historical concern was nephrogenic systemic fibrosis, a rare and serious condition linked to older gadolinium agents in people with severe kidney impairment. Newer macrocyclic agents carry a far lower risk, but screening in this group remains standard.
What this means for you in the waiting room: a handheld analyser takes a drop of blood, usually from a fingerstick, and returns a creatinine and an estimated eGFR in around two minutes. It is a screening tool, not a precision instrument, and it is being used to answer a single binary question about whether your filtration is low enough to change the plan. A borderline point-of-care value is often followed by a proper venous sample rather than acted on directly.
Preparing so the result reflects your actual baseline
You cannot make your kidneys work better before a test, and you should not try. What you can do is remove the noise, so the number you get back means something.
Skip the heavy training for 48 hours. Intense exercise raises muscle turnover and creatinine production for a day or two, and the dehydration that goes with it compounds the effect. This is the single largest avoidable distortion in fit young adults.
Avoid a large cooked-meat meal the evening before. Cooking converts creatine in meat to creatinine, which is absorbed directly. You do not need to go vegetarian; just do not schedule the roast for the night before.
Drink water normally. Being mildly dehydrated reduces blood flow to the kidney and lifts creatinine measurably. Normal hydration, not aggressive water loading, which achieves nothing and can cause its own problems.
Take your usual medicines unless told otherwise. Skipping a dose to make a number look better distorts the picture your doctor is trying to build and risks real harm. If a medicine genuinely needs holding, you will be told explicitly.
Mention creatine supplements, protein powders and any recent antibiotic course. Trimethoprim and cimetidine both raise measured creatinine without affecting filtration at all, by blocking its secretion into urine. A clinician who knows about them expects the bump.
Be consistent about timing if you are being monitored. Creatinine drifts slightly across the day, generally a little higher in the afternoon. If you are tracking a trend across months, having your blood taken at roughly the same time each visit removes one variable.
None of these are about gaming the result. They are about comparability. A creatinine is far more useful as the latest point on a line than as a standalone figure, and every uncontrolled variable makes that line noisier.
Why the same person gets different numbers
Two results from the same person a fortnight apart will rarely be identical, and the difference is often larger than people assume it should be.
Two separate sources of variation stack. Analytical variation is the imprecision of the measurement itself, which on a modern analyser is small — typically a few per cent. Biological variation is the genuine fluctuation in your own creatinine from day to day around your personal set point, which for creatinine sits in the region of 4 to 6 per cent. Combine them and a change of roughly 10 to 15 per cent can occur between two tests with nothing whatsoever having happened to your kidneys.
In practical terms, at a value near 1.0 mg/dL that means a shift of about 0.1 mg/dL, or around 8 to 10 µmol/L, sits inside the noise. This is exactly why a mildly abnormal result is repeated rather than acted upon, and why a diagnosis of chronic kidney disease requires abnormality persisting for at least three months rather than on a single day.
On top of that sits inter-laboratory variation. Even after IDMS standardisation, two laboratories using different methods and different calibrators can return slightly different values for the same sample, and they will certainly quote different reference ranges. If you are comparing results from a hospital, a private clinic and a direct-to-consumer test, some of the difference between them is the laboratory rather than you. Where possible, track your trend within one laboratory.
The rule of thumb clinicians use: a single small change means little, a consistent direction over three or more results means something, and a large sudden change means investigate now. What causes high creatinine covers what to look for in that last case.
What happens after an abnormal result
The sequence is fairly predictable and rarely dramatic. Knowing it in advance removes most of the fear of the unknown.
Acute or chronic is the first question, and old results answer it faster than any new test. A value that has been the same for five years is a different situation from one that doubled since March.
Usually after one to two weeks for a mild elevation, and much sooner if the rise was sharp or you are unwell. A large proportion of single abnormal results are normal on repeat.
New drugs, a recent illness with vomiting or diarrhoea, a contrast scan, unusual exercise, supplements. This step alone explains a substantial share of raised results.
Albumin-to-creatinine ratio and a dipstick for blood and protein. This is where damage that blood creatinine cannot see becomes visible, and it is the step most often skipped by patients who do not realise it was requested.
Quick, painless, no radiation. It rules obstruction in or out decisively and shows kidney size, scarring and cysts.
A nephrology referral follows for markedly reduced or rapidly falling filtration, significant protein in the urine, unexplained findings, or suspicion of a specific kidney disease. Most people with a mildly raised creatinine are managed entirely in primary care.
Occasionally a cystatin C test is added. Cystatin C is a different filtration marker, produced by all nucleated cells rather than by muscle, so it is not thrown off by muscle mass, diet or creatine supplements. It is more expensive and less widely available, which is why it is used as a tiebreaker rather than routinely — typically when the creatinine-based estimate does not fit the clinical picture, or in someone with unusually high or low muscle mass. What a low creatinine means deals with the low-muscle-mass end of that problem.
Red flags that should not wait for a routine follow-up appointment, whatever the number says: a marked drop in how much urine you are passing, new swelling of the legs, ankles or face, breathlessness at rest or lying flat, confusion or unusual drowsiness, or persistent vomiting. Those warrant same-day medical assessment.
The other creatinine tests you may be offered
Blood creatinine is the common one, but it is not the only way creatinine gets measured, and the names are similar enough to cause real confusion.
Urine creatinine (spot sample)
Rarely interpreted alone. Its main job is to normalise other urine measurements for how dilute the sample was — the albumin-to-creatinine ratio and protein-to-creatinine ratio both work this way. A concentrated urine has more of everything, so dividing by creatinine cancels that out.
24-hour urine creatinine clearance
You collect every drop of urine for a full day, and a blood sample is taken during the period. Comparing the two gives a measured rather than estimated clearance. Accurate when done correctly, but incomplete collection is common and quietly ruins the result.
Estimated GFR
Not a separate test. A calculation performed on your blood creatinine using your age and sex, via the CKD-EPI 2021 equation in most laboratories now. The 2021 revision removed the race coefficient that earlier versions applied.
Cystatin C
An alternative blood marker of filtration, independent of muscle mass. Used to clarify borderline or discordant creatinine-based estimates, and increasingly recommended for confirming CKD staging where the decision matters.
If you were handed a large brown bottle and told to collect for 24 hours, that is the clearance test, and the instructions matter more than people realise. Discard the first morning void, collect everything after it including the first void of the following morning, keep it cool, and do not skip one because you were out. A missed sample makes your kidneys look worse than they are.
Cost, access, and self-ordered testing
This varies enormously by country and by insurance arrangement, so specifics here would be useless within a year. The general shape is stable enough to describe.
Where healthcare is publicly funded, as in the United Kingdom, a creatinine requested by your GP or in hospital carries no cost to you at the point of use. In insurance-based systems, it is almost always ordered as part of a panel rather than individually, and the panel price is what appears on any bill. Because creatinine is one of the cheapest analytes to run — a fraction of a millilitre of reagent and a couple of minutes of analyser time — it is rarely the component driving the cost.
Direct-to-consumer testing has expanded considerably, and a creatinine with an eGFR is included in most general wellness panels. Two cautions are worth stating. First, you will receive a number and a range with no clinical interpretation, and the interpretation is where the value lies. Second, results from a consumer laboratory may not be directly comparable with your medical records if a different method or calibration is used, which undermines the trend that matters most. If a self-ordered test comes back abnormal, the sensible next step is to take it to a doctor and expect the test to be repeated through their own laboratory rather than to act on it directly.
The MedlinePlus overview of the creatinine test is a good plain-language reference for what the test involves and what results indicate, written for patients rather than clinicians.
Enter your result in the creatinine clearance calculator for an estimate based on the Cockcroft-Gault equation, then compare it against the normal clearance range for your age. More tools sit in the health calculators library.
Related reading across the cluster: what creatinine is, what high creatinine means, what low creatinine means, what causes high creatinine levels, when to worry about creatinine levels, and what creatinine clearance means.
Creatinine blood tests: frequently asked questions
What is creatinine in a blood test?
Creatinine is a waste product your muscles produce continuously as they use creatine for energy. Healthy kidneys filter it out of the blood into urine at a steady rate, so the concentration left circulating reflects how well filtration is working. The blood test measures that concentration, usually in milligrams per decilitre or micromoles per litre. Because production is fairly constant in any one person, a rising level generally means the kidneys are clearing less. It appears on reports as Creatinine, Serum, Creat, Cr or SCr.
Do you have to fast for a creatinine blood test?
No, creatinine itself does not require fasting, and you can drink water freely. The confusion arises because creatinine is almost always bundled with other tests, and some of those do need a fast — glucose within a comprehensive metabolic panel being the usual one. Follow whatever instruction you were given for the panel as a whole. What genuinely affects a creatinine result is a large cooked-meat meal in the preceding twelve hours, heavy exercise in the previous 48 hours, and turning up dehydrated. Avoid those instead.
Does a basic metabolic panel include creatinine?
Yes. The basic metabolic panel is eight tests and creatinine is one of them, alongside glucose, calcium, sodium, potassium, chloride, bicarbonate and blood urea nitrogen. The comprehensive metabolic panel contains all eight plus albumin, total protein and four liver tests, so it includes creatinine too. A renal function panel adds phosphate and albumin to the basic set. In the United Kingdom the equivalent request is urea and electrolytes, which contains sodium, potassium, urea, creatinine and usually an automatically calculated eGFR.
How is creatinine measured in the laboratory?
Two methods are in routine use. The Jaffe reaction mixes serum with alkaline picrate to produce a coloured complex whose intensity is measured; it is cheap and fast but also reacts with other blood substances, which can nudge results upward. Enzymatic assays use a chain of enzymes specific to creatinine and are more accurate, particularly at low values and in children, but cost more. Both are now calibrated against isotope dilution mass spectrometry, the reference method, so that results are comparable between laboratories.
What does creatinine mean on blood work?
It is the line that reports kidney filtration. A result inside the reference range suggests filtration is adequate, though it cannot rule out early kidney damage, which is why a urine albumin test is usually done alongside. A result above the range means creatinine is accumulating faster than it is cleared, which may reflect reduced filtration or may simply reflect high muscle mass, a recent meat meal, hard exercise, dehydration or certain medications. The eGFR printed beneath it is calculated from the same number.
How do you convert creatinine from µmol/L to mg/dL?
Divide by 88.4. To go the other way, multiply mg/dL by 88.4. That factor comes from creatinine’s molar mass of 113.12 grams per mole worked through the unit change, so it never varies. A value of 88 µmol/L on a British report is the same as 1.0 mg/dL on an American one, and 133 µmol/L equals about 1.5 mg/dL. Knowing this prevents the common alarm of comparing your own report against guidance written in the other unit system.
How long do creatinine blood test results take?
The analysis itself takes minutes. In a hospital emergency department you would typically have a result within 30 to 60 minutes. In primary care it usually takes one to five working days, and the delay is transport, batching and the clinician review step rather than the laboratory work. Point-of-care handheld devices used before contrast scans return a creatinine and eGFR from a fingerstick in about two minutes. If your patient portal releases results before review, wait for the interpretation before drawing conclusions.
What is a normal creatinine level on a blood test?
Approximately 0.7 to 1.3 mg/dL, about 62 to 115 µmol/L, in adult men, and approximately 0.6 to 1.1 mg/dL, about 53 to 97 µmol/L, in adult women. The difference reflects muscle mass rather than kidney function. Ranges vary between laboratories and methods, so use the one printed on your own report rather than any figure from an article. Values are considerably lower in children and drift with age and body composition, and your own previous results are a more useful comparison than any population range.
Why is my creatinine different at two different labs?
Several reasons stack. Your own creatinine fluctuates by roughly 4 to 6 per cent day to day around your personal set point. The measurement adds a few per cent of imprecision. Different laboratories use different methods, calibrators and reference ranges even after international standardisation. Together, a difference of 10 to 15 per cent between two results can occur with nothing having changed in your kidneys. This is why mild abnormalities are repeated, and why tracking your trend within one laboratory is more informative than comparing across several.
What happens if my creatinine test comes back high?
Usually it is repeated, often after a week or two, because single mildly raised results are common and frequently normal on retesting. Your medications, recent illnesses, exercise and diet are reviewed, since these explain a large share of elevations. A urine sample is checked for albumin, protein and blood, because that detects damage a blood test misses. An ultrasound follows if the cause is unclear. Referral to a kidney specialist is reserved for markedly reduced or rapidly falling function, significant proteinuria, or unexplained findings.
The short version
Creatinine on a blood test is a muscle waste product measured to estimate kidney filtration. It is included in the basic metabolic panel, the comprehensive metabolic panel, the renal function panel and the British urea and electrolytes request, so most people get it without asking for it. You do not need to fast for creatinine itself, though other tests on the same form may require it. Avoid a large cooked-meat meal and heavy exercise beforehand, and arrive normally hydrated.
The laboratory measures it by the Jaffe reaction or an enzymatic assay, both standardised against mass spectrometry. Read the result against the range printed on your own report, remember that mg/dL × 88.4 gives µmol/L, and treat the H flag as a prompt rather than a diagnosis. Trend matters more than a single value. Put your number in context with the CrCl calculator, and read further in the creatinine blog category, the wider health blog, or the full tool library at waldev.com.
Medical disclaimer: This article is general educational information about a laboratory test and does not interpret your individual result. It is not medical advice and must not be used to decide whether to seek care, delay care, or change any treatment or medication. Reference ranges and assay methods vary between laboratories, and any result must be read alongside your history, medications, symptoms and other tests. Always discuss your own results 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 a creatinine test involves, why it is ordered and what results mean, in plain language. Creatinine test →
NIDDK on the blood and urine tests used together to assess kidney disease, and how each is interpreted. CKD tests & diagnosis →
The National Kidney Foundation on eGFR, how it is calculated from creatinine, and how the stages are defined. Estimated GFR explained →