Estimates are estimates
Do you know what equation your lab uses to estimate your GFR?
I’ll go back to basics with some renal physiology and jargon explanation and then we can look at the equations. Don’t worry, there is no maths involved.
GFR = glomerular filtration rate
So you need to know that a glomerulus (plural: glomeruli) is the filter part of the kidney. (Actually you will find glomeruli in the cerebellum and the olfactory nucleus in the brain as well, but we wont go there). The glomerulus can be thought of as the first processing station where the blood is filtered, like a sieve that allows waste chemicals and some liquid to wash it through but keeps the important bits in the blood. The filtration rate is the volume of blood filtered in a minute.
This GFR is used a measure of how well the kidney is working: the higher the filtration rate then the better the kidney function.
It is somewhat difficult and actually quite invasive to measure this GFR directly – to do so you have to inject a chemical into the blood and see how quickly it is filtered out of the blood. This can be done using ‘inulin’ (please note I haven’t spelled it incorrectly, inulin is a soluble fibre carbohydrate, not at all related to the hormone insulin). Inulin clearance is seen as the gold standard test for measuring GFR, but it isn’t used clinically very often.
Another way of measuring GFR is using creatinine. Creatinine is a waste product of metabolism so doesnt need to be injected, but the way it is dealt with in the kidney means it doesn’t give such accurate results. Creatinine is filtered by the glomerulus but it is also secreted into the urine further down the nephron. Creatinine clearance does give an idea of kidney function but there are better ways of doing this.
Hence we use estimated GFR or eGFR.
You may come across mGFR for measured GFR, but mostly, in blood test results, the GFR mentioned will be estimated.
Way way back (when I was struggling to get my brain round the nephron in medical school) we used serum creatinine levels to tell us how the kidney was plodding along. That is, we didn’t look its clearance, just used the amount in the blood at the time of the blood test. Obviously there are problems – what if I have just eaten steak or run a marathon (however unlikely)? In those situations the creatinine will be unusually higher so is not really telling me just about kidney function. It is also a problem for old people with no muscle mass, body builders with rather too much muscle mass and if we look at children then it isn’t helpful at all.
Fast forward to 1999: a group of scientists devised an equation that used the creatinine level but also took into account factors of age, gender and ethnicity. This they called the MDRD equation. (MDRD stands for modification of diet in renal disease and this was a study group doing lots of research in the 90s)
The MDRD equation is: …….
OK, I, promised no maths
You can look it up. Its complicated.
For the past decade or so this equation was the standard way to define whether someone had impaired kidney function. The stages of chronic kidney disease were structured around eGFR measurements made using this equation.
Jump to 2009 and another group of scientists – the chronic kidney disease epidemiological collaboration or CKD-EPI for short – came up with a more precise formula. Actually it looks like some of the scientists were in both groups. They devised this new equation from looking at over 8000 patients, using 2/3 of them to work out an equation and validating it on the remaining 1/3. They have added more precise factors to adjust for age and gender and validated the equation using a very large range of people. Ongoing studies are looking at how accurate it is given additional morbidity such as in diabetics,in the elderly, in people after transplant, where there is liver failure etc.
I wont go back on my promise, if you are into maths then feel free to look it up.
To get to the point: the laboratory testing your blood may be using either of these equations.
This would be fine if they came up with the same answer, but in fact the MDRD may give a lower eGFR than the CKD-EPI method and at the extremes, if you are very young, very old or have very low renal function, then the differences between the two formulae may be considerable.
Obviously one laboratory should be consistent in which it uses but for those of us who have health care in different hospitals or move home, or even move country, then there may be a change in your eGFR that does not reflect an actual change in your kidney function but does reflect a change in the method used to estimate it.
Guidelines in UK state laboratories should now be using CKD-EPI. But the lab that just tested my blood only changed to this method last week. In addition I have used three different hospitals for the last three readings and last year I lived in a different country entirely. This means I really don’t know how to interpret the recent changes in my eGFR.
To make things even more complicated, the CKD-EPI people have made further revisions and come up with two more formulae, one using cystatin C instead of creatinine and one using both. Then if you are in China they use C-MDRD also known as the Ma equation, or one called MacIsaac. In US you may also come across the Mayo Quadratic, though less frequently used clinically, and if you are looking at a child’s renal function you will use the Schwartz calculation.
I think the take home messages are that eGFR is an estimate and that to make comparisons it helps to know which equation was used to estimate the estimate.
And I kept my promise: no maths!
Meanings of words:
Nephron = a unit in the kidney that does all the work, 1 kidney contains about 1 million nephrons
Glomerulus = the filter unit of the kidney at the start of each nephron
Creatinine = waste product from broken down protein (not the same as creatine, which is the muscle fuel – I know, its confusing, but the scientists who named them pretended to be classicists by using Greek and lacked imagination and a degree of common sense)
Cystatin C = a protein secreted by most cells in the body that is freely filtered by the glomerulus (later reabsorbed and broken down some other way)
GFR = glomerular filtration rate
MDRD = a group of scientists investigating modification of diet in renal disease
CKD-EPI = a group of epidemiological researchers into chronic kidney disease
Inker, L., Schmid, C., Tighiouart, H., Eckfeldt, J., Feldman, H., Greene, T., Kusek, J., Manzi, J., Van Lente, F., Zhang, Y., Coresh, J. and Levey, A. (2012). Estimating Glomerular Filtration Rate from Serum Creatinine and Cystatin C. New England Journal of Medicine, [online] 367(1), pp.20-29. Available at: http://dx.doi.org/10.1056/nejmoa1114248 [Accessed 7 Oct. 2016].
Levey, A. (1999). A More Accurate Method To Estimate Glomerular Filtration Rate from Serum Creatinine: A New Prediction Equation. Annals of Internal Medicine, [online] 130(6), p.461. Available at: http://dx.doi.org/10.7326/0003-4819-130-6-199903160-00002 [Accessed 7 Oct. 2016].
Levey, A. and Stevens, L. (2010). Estimating GFR Using the CKD Epidemiology Collaboration (CKD-EPI) Creatinine Equation: More Accurate GFR Estimates, Lower CKD Prevalence Estimates, and Better Risk Predictions. American Journal of Kidney Diseases, [online] 55(4), pp.622-627. Available at: http://dx.doi.org/10.1053/j.ajkd.2010.02.337 [Accessed 7 Oct. 2016].
van den Brand, J., van Boekel, G., Willems, H., Kiemeney, L., den Heijer, M. and Wetzels, J. (2011). Introduction of the CKD-EPI equation to estimate glomerular filtration rate in a Caucasian population. Nephrology Dialysis Transplantation, [online] 26(10), pp.3176-3181. Available at: http://dx.doi.org/10.1093/ndt/gfr003 [Accessed 7 Oct. 2016].