Blog | Tuesday, February 26, 2013

New definitions for CKD! Medrants version 1.0

This represents my first attempt at explaining the new CKD definitions. I invite my readers, especially my loyal renal readers, to suggest modifications. This rant will become the basis for a regular talk, and I want to get it right. Thanks in advance for your suggestions.

Ten years after we have new definitions for CKD. Soon after they established the initial stages, authors began to argue that we should divide stage 3 into 3a and 3b. Now they have.

For those who want to read all the details: KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease.

Here is my synopsis of the diagnosis and definitions for CKD:

In order to make a diagnosis of CKD the patient should have at least one of the following:
1) Have an estimated GFR (eGFR) < 60 for 3 months
2) Clear evidence of kidney disease =
--Albuminuria (AER Z30 mg/24 hours; ACR Z30 mg/g [Z3 mg/mmol])
--Urine sediment abnormalities
--Electrolyte and other abnormalities due to tubular disorders
--Abnormalities detected by histology
--Structural abnormalities detected by imaging
--History of kidney transplantation

What are the main points here?
Do not diagnose CKD for patients admitted to the hospital with eGFR 75 unless they have clear evidence of kidney disease. Do not diagnose CKD until you have excluded acute kidney injury (AKI). Patients with a transient creatinine elevation (for example from obstruction or volume contraction) do not have CKD, unless 3 months elapse with continued decreased eGFR.

How should we estimate eGFR?
KDIGO prefers the CKD-EPI formula. A recent JAMA article concluded: "The CKD-EPI equation classified fewer individuals as having CKD and more accurately categorized the risk for mortality and ESRD than did the MDRD Study equation across a broad range of populations." Comparison of Risk Prediction Using the CKD-EPI Equation and the MDRD Study Equation for Estimated Glomerular Filtration Rate

What do all these initials mean? We currently have three creatinine based formulae for estimating GFR. Cockcroft-Gault predates the two more recent models. It uses age, weight, gender and creatinine for estimates. The weight portion does lead to some challenges – ideal weight or actually weight. MDRD comes from the Modifying Diet in Renal Disease study. It uses age, gender, race and creatinine. The more recent CKD-Epi model uses the same variables as MDRD. Most labs still use MDRD.

Here is my favorite spot for renal calculations: eGFR.

When do the formulae not work?
Each formula works by estimating the numerator in the creatinine clearance formula: UV/P. Ucr * V gives the total production of creatinine in 24 hours. Each formula estimates that production, but really is estimating muscle mass, as that is the main source of creatinine. Therefore, if a formula either markedly underestimates or overestimates muscle mass, then the formula will fail. Here are my cautions, i.e., when I eschew any formula:
1) Excess muscle mass: these formulae likely would underestimate GFR in elite body builders or some professional athletes with incredible muscle mass
2) Decreased muscle mass:
--Muscular dystrophies
--Major spinal cord injuries
--Major amputations
--Anorexia nervosa patients

How have the classifications changed?
The old classification had 5 levels:
Stage -- eGFR
1-- >90
2 -- 60-89
3 -- 30-59
4 -- 15-29
5 -- <15

But important epidemiological analyses made clear that stage 3 was too broad. The new staging system starts by dividing stage 3 into 3a and 3b. 3a (45-59) patients have a much lower burden of renal associated complications than do 3b patients:
Stage -- eGFR
1 -- >90
2 -- 60-89
3A -- 45-59
3B -- 30-44
4 -- 15-29
5 -- <15

They go on further to combine the eGFR stages with albuminuria staging. Since I particularly use urine protein/creatinine ratio (PCR) I will produce that chart:
A1 = PCR = <0.15
A2 = PCR = 0.15 – 0.50
A3 = PCR = > 0.50

The higher the A level the more closely we should follow the patient and the more aggressively we should work to delay progression to end-stage.

How and when should we label a patient with the diagnosis CKD?

We should become conservative in applying this label to patients. Once we label the patient they are handicapped in obtaining life insurance, disability insurance, etc.

We should always provide the stage along with the label. Patients do not have CKD, rather the patient has Stage 3b CKD.

The normal estimation equation suffices for 3b or higher, but prior to labeling a patient as having CKD Stage 3a, KDIGO recommends adding a cystatin C measurement and calculating a new estimated GFR. Here is a calculator for this more complex formula.


KDIGO has made staging more complicated, but more useful. The new staging stresses the importance of Stage 3b, which I believe is an advance. It de-emphasizes the risks to Stage 3a patients. The new staging combines proteinuria because proteinuria also predicts complications.

I suspect that it will take significant time to absorb this new staging. I hope this overview will help some of you find this more accessible and more interesting.

db is the nickname for Robert M. Centor, MD, FACP. db stands both for Dr. Bob and da boss. He is an academic general internist at the University of Alabama School of Medicine, and is the Associate Dean for the Huntsville Regional Medical Campus of UASOM. He also serves as a frequent ward attending at the Birmingham VA Hospital. This post originally appeared at his blog, db's Medical Rants.