Creatinine - US vs. rest of the world

Hi Dr Agar. A question came up about converting various blood measures from US to Int’l, for instance creatinine measures. One can find conversion factors online but I don’t really understand how it could work, given the guidance I’ve seen. Standard guidance for Creatinine:
Men: 0.2-0.5 mg/dl (USA) 15-40 umol/L (International)

So looking at this example the low end range for guys is either .2 mg/dl (USA) or 15 umol/L (International), which implies a conversion factor of 75 (.2 * 75 = 15) But the high end implies a conversion factor of 80 (.5 x 80 = 40)

And then the conversion table gives 88.4 as the conversion factor. Where am I going wrong?

Dear Bill

Comparability of method … herein lies the issue!

At the end of this answer, I will copy in an abstract from Peake and Whiting from Clinical Biochemical Reviews which ‘nubs’ the problem … it appeared (I think) about the time when Australia first moved to standardise a national and universal eGFR reporting process for all laboratories in 2007.

What it shows is that there are many methods and/or instruments to measure creatinine - and that the labs often use these different methods or instruments … each with different ‘normals’ and, of course, therefore different conversion scales.

In March 2005, Miller et al published a paper in Arch Pathol Lab Med. Vol 129 titled "Creatinine Measurement: State of the Art in Accuracy and Inter-laboratory Harmonization" [for abstract see<297%3ACMSOTA>2.0.CO%3B2 ] and compared the “results for creatinine measurement from a PT survey of 5624 laboratories using 50 different analysis instruments and analysing method combinations” … 50!

Though there were differences in the reference ranges and accuracy for both (1) the ‘chemical’ method used (there are several) and (2) the analysing machine the method was used in, it appeared that the analysing equipment (the machine) was more of an influence on the final result and accuracy of the creatinine measurement than the chemical method (technique) the laboratory used.

Maybe now you can now see the problem! Each measurement technique (and it’s cost) chosen and each laboratory choice (and affordability) of analysing machine will slightly alter the end result - both in absolute value and degree of accuracy … and sometimes the difference can be quite significantly and yield quite different results.

The same problem of result incomparability will clearly apply, whether laboratory measures and expresses the creatinine result in SI units (umol/L) - as are used in every country in the world except the US … or whether the laboratory expresses its results using the the older way of unit expression (mg/dl) - as is still, inexplicably to me, the case in the US.

Regardless of the units (SI or older version) - each of the different methods will yield a different result and will potentially have a different ‘normal’ value … and, thus, a different conversion figure. Some will differ marginally and be of little or no consequence. Others may be of more obvious discordance.

Moreover, measure the same sample of blood - yes, the same sample, 10 times on the same analytical machine and using the same chemical method - and there will be an element of ‘analytical and method’ error in the results. Not all will be identical. What this tells us that no scienticfic measurement is without an error bar! … an ‘acceptable’ variation from the true mean actual reading. The actual creatinine might be, say, 100 umol/l. But, measure that same sample a few times, and there might be a spread 1 or 2 umol/l each side … ie: 98-102 round the actual mean of 100. This means 98 could also be read as 102 on the same blood sample when the actual reading is 100. I hope that makes sense to you!

This capacity for error … (1) in the chemical method, (2) in the analytical machine and then (3) in the simple inescapable error variation of scientific accuracy … all complicate the picture.

Australian labs are now all standardised and calibrated to use the same method … and that makes it easy here. I cannot speak for the US … but if you read the copied segment from the Miller paper (above), there were 5624 labs in the US using 50 different methods in 2005/6.

The conversion I use for our lab SI units here when converting to the US mg/dl system of reporting is to divide the Australian umol/l figure by 88.4.

… for example, Australian creatinines of 100, 200, 300, 400, 500 and 600 umol/l would, in the US, be respectively 1.13, 2.26, 3.39, 4.52, 5.656, 6.787 mg/dl.

If a lab in the US is using a different method to the standardised Australian measurement system, then the conversion would, admittedly, be different. One really has to know the measuring method for each individual lab to return a true conversion rate.

I think, though, that US mg/dl level multiplied by 88.4 to give a ‘rest of the world’ ummol/l level … or a 'rest of the world ummol/l level divided by 88.4 to give a US mg/dl level … won’t be too far from the truth - at least for most labs in the US.

For further information, see the abstract I referred to at the start … it is copied below …

[I]Peake M and Whiting M. Measurement of Serum Creatinine – Current Status and Future Goals. Clin Biochem Rev. 2006 November; 27(4): 173–184.

Abstract … The first methods for the measurement of creatinine in serum and plasma were published over a century ago. Today, the Jaffe reaction using alkaline picrate remains the cornerstone of most current routine methods, after continuous refinements attempting to overcome inherent analytical interferences and limitations. With the recent introduction of the reporting of estimated glomerular filtration rate (eGFR), inter-laboratory agreement of serum creatinine results has become an important international priority. Expert professional bodies have recommended that all creatinine methods should become traceable to a reference method based on isotope dilution-mass spectrometry (IDMS).

It is important that clinical biochemists have a good understanding of the relative performance of routine creatinine methods. Using a new commutable IDMS-traceable reference material (SRM 967), and a validated tandem IDMS assay developed in our laboratory, we assessed the accuracy of nine routine creatinine methods with assistance from other laboratories in our region. Three methods appeared to have patient sample bias that exceeded 5% in the range of creatinine concentrations where eGFR estimations are most important.

Companies are currently recalibrating their creatinine assays. This task should be complete in 2007, and then creatinine results for eGFR calculations will require the use of a modified eGFR equation. Laboratories considering calibration changes before this time can seek advice from the Australasian Creatinine Working Group. [/I]
Hope that helps and that I havent made it even muddier … but, muddy is as muddy does!

PS: One last comment … all of the above is NOT restricted to creatinine! Most biochemical substances can be measured by a range of units and methods … so it’s a minefield! Sorry to make it all harder … but it truly is a matter of know your lab, know the assay method it uses, and know the method used by the lab that has produced the level that you are then comparing it to’ so you can then know the right conversion to use!’

For those who want to understand more about this … there is a nice paper on this in NDT Plus from earlier this year …

Creatinine determination according to Jaffe—what does it stand for?

Joris R. Delanghe and Marijn M. Speeckaert
Department of Clinical Chemistry, Ghent University Hospital, Gent, Belgium

NDT Plus, (2011) 4: 83-86.


In 1886, Max Jaffe discovered a reaction of creatinine with picric acid in an alkaline environment. Although the manuscript describes the nature of a precipitate and does not deal with an analytical assay, Jaffe’s landmark paper elucidated the basic principles of the creatinine determination method (originally developed by Otto Folin), which became immensely popular and has easily withstood the test of time. Despite the advent of the enzymatic creatinine analysis, the analytical method is still popular due to its simplicity and low cost. As there is no standard recipe for the ‘Jaffe’ method, much methodological variation has occurred over time. This lack of methodological standardization implies that even in the 21st century, improving the interchangeability of Jaffe results is still an issue.