How to properly interpret the creatinine (Cr)

The house officer wades through a swamp of daily creatinines. Unfortunately there is poor example setting on what to do with these. The classic example is the 90 year old with a “normal” creatinine of 90.

Nobody likes formulae, but it is important to refer to a couple here in order to understand what is to come.

Cr Clearance = (Urine volume x urine concentration of Cr) / Plasma Cr

The exact formula is irrelevant for your purposes but it is important to take home the concept that from this formula we can say that Cr clearance is inversely proportional to serum Cr.

Of course we don’t calculate Cr clearance by taking samples of everybody’s piss. That would be far too unwieldy. Instead we estimate Cr clearance using formulae, such as Cockgrauft- Gault, which takes sex, age and weight and spits out a result.

Of course it would be foolish to think we can accurately determine someone’s muscle mass and rate of Cr production even with fancy maths, so these formulae are estimates only, especially at the extremes of age and weight.

What this really boils down to is that I’m more likely to win the lottery than a 90 year old is to have “normal” renal function with a Cr of 90. This, at least, is commonly accepted, although commonly ignored, probably because we all have a habit of only looking at the exact number if it appears in red.

The nephron deepens

There are more interesting conclusions we can come to just from looking at the Cr clearance formula. Consider the graph that plots the function y = 1/x (CrCl being proportional to 1 / Cr)


If the y axis is Cr clearance and the x axis is Cr, what you can see is that

  • On the first part of the graph a fairly significant fall in Cr clearance is accompanied by only a small increase in Cr
  • Towards the end of the graph a fairly small drop off in Cr clearance is accompanied by a large increase in Cr

This is extremely important. A rise in Cr (even in a young healthy person) from 50 to 80 represents a significant kidney injury. Such changes in Cr are mostly ignored, because it is all ‘in the black’. In addition, Cr is usually stable from measurement to measurement in a person, and the variability in the lab assay is low. It is really a mistake to think of Cr as having a reference range- the reference range is what is normal for that person and even apparently small variations are important.

Conversely a rise in Cr from 450 to 500 will catch everybody’s attention but represents a much smaller further deterioration (that’s not to say you shouldn’t act on it). Or more importantly, when the Cr falls from 500 to 450, more cheers erupt than are deserved.

It is important to mention here that despite the relationship between acute rises in Cr and the fall in Cr clearance, it is impossible to actually determine what someone’s Cr clearance is when they have an acute kidney injury. The formulae are all designed for the situation when Cr production = Cr excretion. Obviously in an AKI when the Cr is rising this is not the case and the formulae cannot be used.

This means that it is inappropriate for the radiologist to ask you to plug in the patient’s Cr into the Cr clearance calculator, get a number that says 31ml/min, and say “yep we can go ahead and give contrast” (this happens all the time). The Cr clearance will be much worse than the number the calculator spits out (this is not to mention that the cutoff of 30ml/min is fairly arbitrary anyway). In any situation such as this it is important to get a higher up to make the decision that the diagnostic benefits of contrast outweigh the risks of kidney injury.

A false rise in Cr

The clearance of creatinine is a good estimate for glomerular function because it is mostly filtered at the glomerulus and only some is secreted in the tubules. However there are some drugs that will inhibit tubular secretion and raise the Cr, even though renal function has stayed the same. The main drugs that do this are cotrimoxazole and cimetidine. Sometimes you will see this referred to as a “cotrimoxazole induced AKI”. This is not really correct as no AKI has actually taken place.

The effect of fluid balance

Cr, like any molecule that resides in the plasma, will be diluted when the extracellular water compartment increases, and concentrated when it decreases. Imagine the post surgical patient who has received 3L of fluid in theatre and an extra few litres from you on the ward. The next morning their Cr has gone from 70 to 80. In actuality the Cr would have risen more if it wasn’t diluted by 5L of salt. Therefore, the more positive the fluid balance, the more the true extent of renal injury has been “masked” by diluting the Cr. In fact you can adjust Cr for fluid balance, although these calculators are rarely used.


The degree of renal injury is often underappreciated. You can avoid this if you look at the numbers themselves rather than normal/abnormal, understand the relationship between Cr and renal function, and understand that estimates of Cr clearance have many limitations and that serum Cr is influenced by fluid balance.


Serum Creatinine and Glomerular Filtration Rate: Perception and Reality. R Neil Dalton. Clinical Chemistry 2010.

Fluid accumulation, recognition and staging of acute kidney injury in critically-ill patients. Macedo, Bouchard and Mehta. Critical Care 2010;14(3):R82

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s