AKI and CKD Staging: The Evidence Behind the Pathway

Two men come to the diabetes clinic on the same morning. Mr Rao is 58, with an eGFR of 92 mL/min/1.73 m² and a urine albumin-to-creatinine ratio of 450 mg/g. Mr Shetty is 64, with an eGFR of 50 and an albumin-to-creatinine ratio of 12 mg/g. The intern looks at the two eGFRs and says, “Mr Shetty’s kidneys are the worry.” These are illustrative cases, not real patients.

By the end of this article you will know why that reading is backwards, and how to read a creatinine that has just gone up.

1. Why CKD Is Staged on Two Axes

“Is the eGFR not the kidney function?” It is half of it. The tool used to explain the other half like this:

CKD was once staged on eGFR alone, which hides the patient who is still filtering normally and is losing the filter while you watch. Albuminuria marks active glomerular endothelial damage and podocyte effacement, so it reports damage happening now, where eGFR reports what has already gone.

So the two axes answer two different questions. The eGFR says how much filter is left. The albuminuria says whether it is being damaged now. A patient with a normal eGFR of 90 (G1) and heavy albuminuria (A3) carries a higher risk of progression to kidney failure and of cardiovascular death than a patient with an eGFR of 50 (G3a) and no albuminuria (A1). That is Mr Rao against Mr Shetty.

2. Reading the KDIGO Heatmap

KDIGO puts the two axes on one grid. Here it is as one table, as the tool used to carry it:

Find each man’s cell. Mr Rao is G1 and A3: high risk. Mr Shetty is G3a and A1: moderate risk. The eGFR that worried the intern is the lower risk of the two.

Two points from KDIGO 2024 change how the grid is read. First, where cystatin C is available, KDIGO asks that the GFR category be assigned from the combined creatinine and cystatin C equation. The tool computes CKD-EPI 2021 from creatinine alone, which is right when cystatin C has not been measured; a stage near a boundary is provisional. Second, KDIGO 2024 moves the emphasis towards individual risk prediction, using the Kidney Failure Risk Equation for the 2 and 5 year risk of kidney failure (Tangri). Two patients in the same cell can carry different absolute risk.

3. A Creatinine That Rises: True AKI or Not?

Three weeks later Mr Shetty is started on cotrimoxazole for a urinary infection. His creatinine goes up by 0.4 mg/dL. His urea is normal, his potassium is normal, and he is passing urine as usual. “Should we stop the antibiotic?”

Not on this alone. Trimethoprim, in cotrimoxazole, blocks the tubular secretion of creatinine at the OCT2 and MATE1 transporters in the proximal tubule. The actual GFR remains normal. The tool calls this pseudo-AKI, and sets it beside true AKI:

• True AKI: a rising creatinine accompanied by a rising urea, electrolyte derangement such as hyperkalaemia, and usually altered urine output. The next step is the urinary indices, FeNa and FeUrea, to separate pre-renal from intrinsic.
• Pseudo-AKI: an isolated rise in creatinine, usually a 0.3 to 0.5 mg/dL bump at most, with a normal urea and no electrolyte derangement, which resolves entirely when the drug is stopped. Cimetidine and cobicistat do the same. Do not stop beneficial therapy or reflexively order dialysis.

The common mistake is to read every creatinine rise as injury. It is tempting, because the creatinine is the number on the report. Look at the urea and the potassium beside it. And one caution: fenofibrate is not one of these. It raises creatinine too, but not by secretion blockade alone, and a genuine fall in measured GFR has been reported on it. A rise on fenofibrate needs a measured clearance rather than reassurance.

4. The Nephrotoxins a Patient Will Not Volunteer

Much of the unexplained acute-on-chronic renal failure seen in Indian practice comes from two sources the patient will not volunteer: over-the-counter NSAIDs, diclofenac and aceclofenac above all, bought without a prescription and taken for months for a bad knee; and unregulated alternative and Ayurvedic preparations, bhasmas among them, some of which carry heavy metals.

“Why are NSAIDs so hard on a kidney that is already struggling?” The tool used to give the reason:

NSAIDs constrict the afferent arteriole, and in a patient whose perfusion already depends on prostaglandin-mediated dilatation that is enough to drop the GFR hard.

Ask about both by name. A patient asked whether he takes any medicines will say no, and will mean the ones a doctor wrote down. Mr Shetty, asked by name, buys aceclofenac at the chemist for his knee.

5. Why Cockcroft-Gault Is Still Printed

Cockcroft-Gault was published in 1976, before creatinine assays were standardised, and it takes raw body weight into its numerator. In an obese or fluid-overloaded patient a good deal of that weight is not muscle, so the equation credits the kidney with a clearance it does not have. KDIGO asks for CKD-EPI when staging disease, and CKD-EPI is the number to stage with.

“Then why does the tool print it at all?” For dosing. The renal dose adjustments on FDA labels for drugs such as vancomycin and digoxin were derived against Cockcroft-Gault. A label that says to reduce the dose below a given clearance means clearance as Cockcroft-Gault measured it. That is the only thing the figure is there for.

Back to Our Patients

Mr Rao, with the better eGFR, is the one at high risk on the heatmap, and his albuminuria is the reason. Mr Shetty is at moderate risk. His creatinine bump on cotrimoxazole, with a normal urea and potassium, reads as pseudo-AKI, and the antibiotic can continue. His aceclofenac stops. For any drug whose label doses by clearance, his Cockcroft-Gault figure is the one to use.

Try This

A 70-year-old woman has an eGFR of 38 and an albumin-to-creatinine ratio of 80 mg/g. Where does she sit on the heatmap? Illustrative case.

Answer. An eGFR of 38 is G3b (30 to 44), and 80 mg/g is A2 (30 to 300). G3b with A2 is very high risk. If her creatinine-only eGFR were close to a boundary, say 45 or 46, a cystatin C would be worth measuring before the stage changes her management.

Take-Home Points

• Stage CKD on both axes: the eGFR says how much filter is left, the albuminuria whether it is being damaged now.
• Normal eGFR with heavy albuminuria can carry more risk than a low eGFR without it.
• An isolated creatinine bump with a normal urea and potassium on trimethoprim is usually pseudo-AKI: do not stop beneficial therapy.
• Ask about over-the-counter NSAIDs and Ayurvedic preparations by name.
• Stage with CKD-EPI; dose by the label’s Cockcroft-Gault clearance.

For a creatinine that is still moving, see Kinetic GFR: The Evidence Behind the Calculator; for the potassium beside it, Acute Hyperkalaemia: The Evidence Behind the Protocol.

Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Evidence tab on 5 November 2026, each from the source the tool names. The illustrative cases follow the tool’s rules; check the tool’s result for a real patient.

References

1. Inker LA, et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385(19):1737-1749.
2. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31-41.
3. Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int. 2024;105(4S):S117-S314. [Current edition; supersedes the 2012 CKD guideline]
4. Stevens PE, Ahmed SB, Carrero JJ, et al. Executive summary of the KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease: known knowns and known unknowns. Kidney Int. 2024;105(4):684-701.
5. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl. 2013;3(1):1-150. [Superseded 2024]
6. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int Suppl. 2012;2(1):1-138.
7. Tangri N, Stevens LA, Griffith J, et al. A predictive model for progression of chronic kidney disease to kidney failure (Kidney Failure Risk Equation). JAMA. 2011;305(15):1553-1559.
8. Indian CKD Guideline Workgroup, Indian Society of Nephrology. Indian Chronic Kidney Disease Guidelines. Indian Society of Nephrology; December 2013.