Understanding the kidney failure risk calculator (KFRE) — concept and principle
The Kidney Failure Risk Equation (KFRE) estimates the chance that an adult with chronic kidney disease (CKD) stage 3–5 will need dialysis or a kidney transplant within 2 or 5 years. This is the four-variable version using age, sex, eGFR and urine albumin-to-creatinine ratio (ACR). It helps decide when to refer to a nephrologist and when to plan for dialysis access or transplantation.
How it is calculated
L = −0.2201 × (age ÷ 10 − 7.036) + 0.2467 × (male − 0.5642) − 0.5567 × (eGFR ÷ 5 − 7.222) + 0.4510 × (ln ACR − 5.137)
Risk at 2 years = 1 − 0.9750^exp(L); at 5 years = 1 − 0.9240^exp(L)
- Male = 1 for men, 0 for women. ACR is in mg/g; mg/mmol × 8.84 = mg/g.
- The baseline survival values are the North American calibration.
The principle
The equation is a Cox proportional hazards model developed by Tangri and colleagues and validated in more than 700,000 people worldwide. The linear predictor L measures how far a person's risk factors are from the average patient in the development data: lower eGFR and higher albuminuria raise it strongly. The baseline survival is the chance that an average patient stays off dialysis; raising it to the power exp(L) adjusts it for the individual.
Worked example
Default inputs: age 65, male, eGFR 35, ACR 300 mg/g.
- Age term = −0.2201 × (6.5 − 7.036) = +0.118.
- Sex term = 0.2467 × (1 − 0.5642) = +0.108.
- eGFR term = −0.5567 × (7 − 7.222) = +0.124.
- ACR term = 0.4510 × (ln 300 − 5.137) = 0.4510 × 0.567 = +0.256.
- L = 0.605; 2-year risk = 1 − 0.975^1.831 = 4.5%; 5-year risk = 1 − 0.924^1.831 = 13.5%.
| Urine albumin/creatinine ratio (ACR) | Kidney failure risk in 2 years |
|---|---|
| 75 | 2.4% |
| 150 | 3.3% |
| 225 | 4% |
| 300 | 4.5% |
| 375 | 5% |
| 450 | 5.4% |
| 600 | 6.1% |
How to read the result
KDIGO 2024 suggests a 5-year risk of 3–5% as a threshold for referral to a nephrologist, a 2-year risk above 10% for multidisciplinary care and above 40% to start preparing for dialysis or transplant; local practice varies. Outside North America a recalibrated version gives somewhat lower risks.
Common mistakes
- Using a protein/creatinine ratio instead of an albumin/creatinine ratio.
- Entering mg/mmol while the unit is set to mg/g.
- Using it with eGFR of 60 or above, where it was not designed.
Frequently asked questions
Does KFRE include diabetes or blood pressure?
No. Their effect works largely through eGFR and albuminuria.
Is there an eight-variable version?
Yes. It adds calcium, phosphate, bicarbonate and albumin.
Does the result predict death?
No. It predicts kidney failure needing dialysis or transplant, not death from other causes.
Sources: Tangri N et al., "A predictive model for progression of chronic kidney disease to kidney failure", JAMA 2011;305(15):1553–1559 · Tangri N et al., "Multinational assessment of accuracy of equations for predicting risk of kidney failure", JAMA 2016;315(2):164–174 · KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease, Kidney Int 2024;105(4S):S117–S314