What Is the Normal Urine Protein/Creatinine Ratio?

The protein-to-creatinine ratio, written PCR or UPCR, takes the amount of protein in a single urine sample and divides it by the amount of creatinine in that same sample. In healthy adults the answer comes out below roughly 15 mg/mmol, or below roughly 150 mg/g, depending on which units your laboratory uses. Above that, protein is leaking into the urine at a rate worth explaining. Above 50 mg/mmol or 500 mg/g, it is worth investigating properly. Above about 300 mg/mmol or 3.5 g/g, it is heavy enough to be called nephrotic-range, and that carries its own set of consequences.

Why divide by creatinine at all? Because urine concentration swings enormously across a day and would otherwise wreck the measurement. Drink a liter of water and your urine protein concentration halves without a single molecule of protein changing hands. Creatinine, by contrast, is excreted at a fairly steady rate hour after hour, so it acts as a built-in yardstick for how dilute the sample was. Dividing one by the other cancels the dilution out. That single trick is what makes a random pot of urine as informative as a 24-hour collection, and it is why almost nobody carries a jug around for a day any more. If you want the mechanics of doing the sum yourself, how to calculate the protein-creatinine ratio walks through it step by step; if your result came back high and you want the reasons, what causes a high protein-creatinine ratio covers those separately.

What the ratio is, and the one reason it exists

Healthy kidneys keep protein where it belongs. The filtering membrane is fine enough that albumin and the other large plasma proteins mostly stay in the blood, and what little slips through is largely reabsorbed further down the tubule. A small amount escapes anyway. Under about 150 mg of protein leaves in the urine over a normal day, much of it not plasma protein at all but a tubular secretion called Tamm-Horsfall protein. That small background leak is why the normal range is not zero.

Measure the protein concentration in a single sample, though, and you learn almost nothing. Concentration depends on how much water the urine has been diluted with, and that varies by a factor of ten or more between a dehydrated morning sample and one passed twenty minutes after a pint of squash. Two samples from the same person on the same day can read 8 mg/dL and 80 mg/dL with identical actual protein leakage.

Creatinine solves this. Your muscles convert creatine to creatinine at a rate that barely changes from hour to hour, and the kidneys clear it steadily, so the quantity arriving in the urine per unit time is close to constant for any given person. That makes urine creatinine a proxy for how much urine the sample represents. High creatinine concentration means a concentrated sample. Low creatinine means a dilute one. Divide protein by creatinine and the dilution factor appears on both sides of the division and vanishes.

What comes out the other side is a rate rather than a concentration: roughly how much protein you lose per unit of creatinine, and since creatinine output per day is fairly predictable, roughly how much protein you lose per day. That is the whole idea. A number that no longer cares whether you had a drink before the appointment.

Creatinine’s steadiness is what makes the whole thing work, and it is worth understanding on its own terms. MedlinePlus sets out how creatinine is measured in both blood and urine, and why it appears on so many different tests for so many different reasons.

Two names for the same test appear on reports and cause needless confusion. UPCR stands for urine protein-creatinine ratio; PCR for protein-creatinine ratio; some labs write “urine protein/creat ratio” or just “prot:creat”. They are the same measurement. PCR occasionally also stands for polymerase chain reaction in a completely unrelated context, which has caused more than one person to search themselves into a panic about the wrong test entirely.

The normal range, in every unit you are likely to meet

Here is the table most people came for. Every row is the same clinical situation expressed in different units, so read across rather than down.

A few things about that table deserve saying out loud, because reference tables tend to be read as more precise than they are.

One more distinction worth fixing early. This test measures total protein, meaning albumin plus everything else, including immunoglobulin light chains and the low-molecular-weight proteins that appear in tubular disease. The albumin-to-creatinine ratio is a different test with much lower thresholds, and mixing the two up is the second commonest error after units. There is a section on that further down, and how to calculate the albumin-creatinine ratio covers it in full.

Converting between mg/mmol, mg/g and g/g

This is the single biggest source of confusion with this test, and it comes down to one number: 8.84.

Creatinine has a molecular mass of 113.12 grams per mole, so one millimole of creatinine weighs 113.12 milligrams, or 0.11312 grams. A result expressed as milligrams of protein per millimole of creatinine is therefore milligrams of protein per 0.11312 grams of creatinine. To express it per whole gram instead, divide by 0.11312, which is the same as multiplying by 8.84.

If you are converting in your head, multiplying or dividing by ten gets you close enough to know which band you are in. A ratio of 40 mg/mmol is somewhere near 400 mg/g; the exact answer is 354, which sits in the same category and leads to the same conversation. Use 8.84 when the number matters and 10 when you just want orientation.

Notice the wobble at the nephrotic end. Some sources define nephrotic-range proteinuria as over 300 mg/mmol, others as over 3.5 g/g, and those two are not quite the same number: 3.5 g/g is 396 mg/mmol. Again, the boundary is a convention that grew up separately in different unit systems. Nobody with a ratio between 300 and 400 mg/mmol is having a subtly different disease depending on which threshold their hospital uses; they are all being investigated.

The bands, from normal to nephrotic

Reference ranges are easier to hold onto as a scale than as a table. Here is the whole span in one strip, with the conventional adult thresholds marked in both common unit systems.

What each band tends to mean in practice, keeping in mind that the number never diagnoses anything on its own:

The threshold that carries most weight in routine practice is 50 mg/mmol, or 500 mg/g. Below it, a raised ratio is usually watched and the underlying risk factors managed. Above it, and particularly if it persists across repeat samples, most guidelines push toward specialist assessment regardless of how good the filtration rate looks. Proteinuria and filtration are separate axes of kidney health, which is why calculating GFR from creatinine answers a different question from this one, and why staging systems use both.

A worked example, in both unit systems

Numbers on a page are easier to trust once you have watched one go through the mill. Take a 46-year-old woman with type 2 diabetes having a routine check. Her first morning sample comes back with a urine protein of 420 mg/L and a urine creatinine of 8.4 mmol/L.

Now the same person, same day, second sample taken two hours after she drank a large glass of water. Protein 168 mg/L, creatinine 3.36 mmol/L. The protein concentration has dropped by 60 percent, which looks like dramatic improvement. Divide: 168 ÷ 3.36 = 50 mg/mmol. Identical. The dilution moved both numbers by the same factor and the ratio did not budge. That is the entire justification for the test in one line of arithmetic.

The American version of the same calculation starts from different units. Suppose a report gives urine protein 46 mg/dL and urine creatinine 118 mg/dL. Divide directly: 46 ÷ 118 = 0.39. Because both are in mg/dL, the units cancel and the answer is already in mg/mg, which is the same as g/g. Multiply by 1000 to get 390 mg/g. Divide that by 8.84 and you get 44 mg/mmol. One result, three ways of writing it, all sitting in the mild band.

How the ratio relates to the old 24-hour collection

For decades the reference standard was a 24-hour urine collection: every drop passed over a full day, kept in a container, measured for total protein. It is still occasionally done. It has largely been displaced by the ratio, and the reason is not that it was inaccurate in principle but that it was inaccurate in practice.

Collections go wrong constantly. People forget a void, include the first morning sample they were supposed to discard, spill some, or quietly give up by teatime and hand in eighteen hours labeled as twenty-four. There is no way for the laboratory to tell. An incomplete collection understates protein loss, sometimes badly, and it does so invisibly. A single pot of urine, by contrast, is difficult to get wrong.

The rough equivalence is easy to hold onto:

So a ratio of 800 mg/g suggests about 800 mg of protein a day. A ratio of 60 mg/mmol suggests about 530 mg a day. The reason this works is the assumption behind it: the average adult excretes close to one gram of creatinine in 24 hours, so protein per gram of creatinine is protein per day.

That assumption is where the error lives. Creatinine excretion tracks muscle mass, and muscle mass varies enormously.

The practical consequence: a bodybuilder with a ratio of 45 mg/mmol may actually be losing considerably more protein per day than the conversion suggests, while an 82-year-old with the same ratio may be losing less. Neither is a reason to distrust the test. It is a reason to treat the “grams per day” translation as an approximation and to follow the ratio itself over time rather than the derived daily figure. Trends in the same person, using the same units and the same collection habit, are what matter. The same logic that applies to muscle mass and normal blood creatinine levels applies here, for the same underlying reason.

There are still situations where a 24-hour collection gets requested: when the ratio and the clinical picture disagree, in some research settings, and occasionally in pregnancy where the stakes of a borderline value are high. But for screening, monitoring and most diagnosis, the ratio has won on practicality alone. The NIDDK guidance on kidney disease testing sets out why urine and blood tests are read together rather than in isolation.

Normal values in children, which are not the adult ones

Children are not small adults for this test, and applying an adult threshold to a toddler produces alarm where none is warranted. The reason is straightforward: young children have very little muscle, so they excrete very little creatinine, so the denominator of the ratio is small and the ratio itself sits higher for the same amount of protein.

A ratio of 0.35 g/g in an eighteen-month-old is within the expected range. The same value in a fifteen-year-old is clearly abnormal. Pediatric laboratories print age-banded ranges for exactly this reason, and a parent comparing a child’s result against a general reference table found online is one of the more common ways to frighten yourself unnecessarily.

One pediatric pattern deserves its own mention because it is common, benign and frequently mishandled: orthostatic, or postural, proteinuria. In some adolescents and young adults, protein appears in the urine when they are upright and disappears when they are lying down. A sample collected during the school day shows a raised ratio. A first morning sample, passed immediately on waking after a night horizontal, comes back normal. The condition is harmless, needs no treatment, and resolves in most people over years. It is also the single best argument for insisting on a first morning specimen before anyone investigates a mildly raised pediatric result.

Persistent proteinuria in a child is a different matter and is taken seriously, because the underlying causes tend to be glomerular disease rather than the slow vascular damage that dominates adult kidney disease. Any raised ratio in a child that persists across two properly collected first morning samples warrants pediatric assessment. So does a raised ratio accompanied by swelling around the eyes or ankles, visible blood in the urine, or high blood pressure.

Normal values in pregnancy, where the threshold moves

Pregnancy changes the numbers, and it changes what they are being used for. Normal pregnancy raises the filtration rate by something like 40 to 50 percent, and more filtration means more protein slipping through even with a perfectly healthy kidney. Baseline protein excretion rises accordingly.

The threshold used in pregnancy is therefore higher than the non-pregnant one, and it is set for a specific purpose: detecting pre-eclampsia.

Twice the non-pregnant limit, and deliberately so. A ratio of 22 mg/mmol at 30 weeks would be flagged in a non-pregnant adult and is below the pregnancy action threshold. This is not laxity; it reflects the physiology, and the threshold was chosen because 300 mg a day is the level at which proteinuria starts to carry meaning in the context of raised blood pressure.

Context is the crucial word. Proteinuria in pregnancy is almost never assessed alone. It is read alongside blood pressure, symptoms, blood tests including platelets, liver enzymes and creatinine, and the baby’s growth. A ratio above 30 mg/mmol with normal blood pressure and no symptoms means something quite different from the same ratio with a blood pressure of 158/104 and a headache. Pre-eclampsia can also occur without significant proteinuria at all, which is why the diagnosis no longer requires it.

Two further points about pregnancy testing. A urinary tract infection raises the ratio and is common in pregnancy, so a positive result is usually checked against a urine culture before it is acted on. And because the consequences of getting a borderline pregnancy result wrong are significant, this is one of the settings where a 24-hour collection is still sometimes requested to settle the question. Blood creatinine behaves differently in pregnancy too, running lower than usual because of the increased filtration; creatinine levels in pregnancy covers that side.

What a borderline result actually means

A ratio of 18 mg/mmol against a reference range topping out at 15. Or 165 mg/g against a limit of 150. This is the commonest result people go looking for answers about, and the honest answer is that a single borderline value means very little on its own.

Protein excretion is not fixed. It rises after exercise, during a fever, with a urinary tract infection, in the days around a period, after a large protein meal, during heart failure, and in some people simply from being upright all day. A great many mildly raised ratios are transient and normalize on a repeat sample. That is why guidelines almost universally ask for confirmation before anything is done, and why a single value at the edge of the range is not treated as a diagnosis of anything.

The standard approach is to repeat the test on an early morning sample, ideally twice over the following weeks, and to check what else is happening at the same time: blood pressure, kidney function on a blood test, and whether there is blood in the urine alongside the protein. That combination sorts the great majority of borderline results. Protein and blood together in the urine of someone with a raised blood pressure is a different situation from isolated mild protein in an otherwise well person, and it is investigated more quickly.

Where a borderline result does turn out to be persistent, it is worth knowing that this is not a small finding dismissed. Even modestly raised protein excretion is one of the better predictors of kidney function declining over the following years, and it is also an independent marker of cardiovascular risk. That is precisely why it gets followed rather than filed. Being told to come back in three months is a reasonable plan, not a brush-off. If your filtration is also in question, creatinine levels in stage 1 kidney disease covers the situation where protein is present but filtration is still normal, which is a real and specific category.

How the sample should be collected for a valid result

The test is only as good as the pot. A ratio calculated from the wrong sample at the wrong time can be wrong by a factor of two or more, and nothing in the laboratory will flag it.

None of that requires fasting, stopping medication or any other preparation. There is no need to restrict fluid, and no benefit in loading up on water beforehand either. The whole point of the ratio is that hydration does not change it, so drinking a lot before the test neither helps nor hurts, though it can dilute the sample far enough to cause a different problem, covered next.

Why very dilute or very concentrated urine distorts the ratio

The ratio corrects for dilution, but only within a working range. Push the sample far enough in either direction and the correction starts to break down.

In a very dilute sample, both protein and creatinine sit close to the assay’s lower limit of detection. Measurement error is roughly constant in absolute terms, so when the numbers themselves are tiny, that error becomes a large proportion of each one. Dividing one imprecise small number by another imprecise small number produces a ratio with a wide margin of uncertainty. Many laboratories flag or decline samples with a urine creatinine below roughly 4 mmol/L, or about 45 mg/dL, for this reason.

There is a second, subtler problem with dilute samples. Some protein assays are less accurate at low concentrations and may under-read, while creatinine measurement holds up slightly better. The net effect can be a ratio that reads falsely low. Someone who drinks two liters of water before giving a sample may hand in a specimen that is essentially uninterpretable, and in the worst case falsely reassuring.

Very concentrated urine has the mirror problem. Creatinine is high, the denominator is large, and the ratio is pulled downward. Overnight samples from someone who is dehydrated can sit at the concentrated end. In practice this matters less than the dilute case, because the assays are comfortably within range and the distortion is modest, but it is a real reason why the same person can produce ratios differing by 20 or 30 percent across samples with no change in their kidneys at all.

The upshot is simple and worth following: drink normally before the test, neither more nor less than usual. What counts as high or low for the creatinine value itself is covered in high urine creatinine and low urine creatinine, and dehydration’s broader effect on creatinine is set out in can dehydration cause high creatinine levels.

PCR and ACR: two tests, two sets of numbers

The albumin-to-creatinine ratio, ACR, uses the same denominator and a narrower numerator. It measures albumin alone rather than total protein, and its thresholds are roughly a tenth of the PCR ones. Reading an ACR result against a PCR reference range, or the reverse, produces a wildly wrong conclusion in either direction.

Which one you get depends on the question. For screening someone with diabetes or hypertension, ACR wins, because early glomerular damage leaks albumin first and in quantities too small for a total protein assay to see reliably. For someone already known to have heavy proteinuria, or where the concern is a protein other than albumin, PCR is the more informative test. Light chains in myeloma and the low-molecular-weight proteins of tubular disease are invisible to an albumin assay and would be missed entirely.

Both can be requested, and in some settings both are. They do not contradict each other; they answer adjacent questions. A rough rule of thumb, worth treating as orientation rather than arithmetic: in albumin-predominant kidney disease, ACR accounts for perhaps 50 to 70 percent of the PCR value, with the remainder made up of other proteins. That relationship falls apart when the leaked protein is not albumin, which is exactly when it is most important to have measured total protein.

When the ratio gives a misleading answer

Every test has failure modes and this one has a specific set, all of which come back to the denominator.

That last row is worth a moment. The whole method assumes creatinine excretion is steady. During acute kidney injury it is not, because creatinine is accumulating in the blood rather than being cleared at a stable rate. A ratio measured in the middle of a rapidly deteriorating episode is difficult to interpret and should not be compared against one taken when things were stable. This is the same steady-state assumption that underpins creatinine clearance and every equation built on it, and it fails in the same circumstances.

None of these makes the test bad. They make it a test that needs its context, like every other. A number read off a report without knowing whether the person is a powerlifter, has a fever, or gave the sample at four in the afternoon after a gym session is a number that can easily be read wrong.

Reading your own lab report, line by line

Reports vary, but most contain the same handful of lines and the same handful of traps.

That last step catches people out more than any other. A ratio of 40 last year reported in mg/mmol and a ratio of 380 this year reported in mg/g look like a tenfold deterioration and are in fact the same result. Before worrying about a jump, check that both numbers are in the same unit. Laboratories do change reporting conventions, and they do not always make it obvious.

What happens after an abnormal result

Knowing the likely sequence takes some of the sting out of being told your urine test was abnormal.

The first move is nearly always confirmation. One raised ratio on a random sample is not enough to act on, so a first morning specimen is requested, often twice, spaced over weeks. A surprising proportion of mildly raised results do not reproduce. If they do not, and there is nothing else abnormal, that is usually the end of it.

If it does reproduce, the assessment broadens rather than deepens. Blood pressure, measured properly. A blood test covering kidney function, glucose and sometimes more. Urine tested for blood as well as protein, because the two together point toward inflammation in the filtering units in a way that protein alone does not. Often an ultrasound to look at kidney size and structure. This is a fairly standard workup and it does not imply anyone expects to find something serious.

Referral to a kidney specialist tends to follow certain triggers rather than a single number: a ratio persistently above 50 mg/mmol or 500 mg/g, protein together with blood in the urine, a falling filtration rate, blood pressure that will not settle, or any nephrotic-range result. A biopsy comes into the conversation when the pattern suggests glomerular disease and the answer would change treatment, which is more often than people expect at the heavier end.

Treatment, when there is any, is usually aimed at two things at once: whatever is causing the leak, and the leak itself. Blood pressure control matters enormously, and certain classes of blood pressure medication reduce protein loss beyond their effect on pressure alone, which is why they are chosen preferentially in people with proteinuria. Diabetes control matters for the same reason. Nothing about any of that is something to start, stop or adjust on your own reading; it is a conversation with the doctor who has your full picture. For general background on protecting function, how to lower creatinine levels and how to prevent creatinine rising cover the lifestyle side, and when to worry about creatinine levels covers the thresholds that matter on the blood side.

Related reading across the cluster: what creatinine is, what high creatinine means, creatinine on a blood test, creatinine clearance versus GFR, and the BUN-to-creatinine ratio, which is a different ratio entirely and frequently confused with this one.

Normal urine protein-creatinine ratio: frequently asked questions