Estimate the glomerular filtration rate (eGFR) from serum creatinine, age and sex using four equations side by side — CKD-EPI 2021 (race-free), CKD-EPI 2009, MDRD and Cockcroft-Gault — and visualize the KDIGO CKD stage (G1-G5), albuminuria category (A1-A3), dialysis threshold and projected years to ESRD in real time.
Parameters
Age
year
Serum creatinine
mg/dL
Serum creatinine — rises as kidney function falls
Sex
Equation
CKD-EPI 2021 is the modern standard (race-free)
Body weight
kg
Used only by Cockcroft-Gault
Race
Used by CKD-EPI 2009 / MDRD only
UACR (urine albumin/Cr)
mg/g
A1: <30 / A2: 30-300 / A3: ≥300
Results
—
eGFR (mL/min/1.73m²)
—
CKD stage
—
Albuminuria category
—
High risk
—
Dialysis needed
—
Years to ESRD
—
Kidney & nephron filtration animation
Blood is filtered in the glomerulus (capillary tuft) and primary urine flows from Bowman's capsule into the proximal tubule. Colour shows the CKD stage (green G1 → orange G3 → red G5).
Cockcroft-Gault (1976). CrCl is in mL/min (not BSA-normalized). FDA drug-label renal dose adjustments are still written predominantly in CrCl terms. W: body weight in kg.
The KDIGO heat map stratifies cardiovascular and dialysis risk on two axes — eGFR and albuminuria (A1 <30, A2 30-300, A3 ≥300 mg/g).
Kidney function eGFR & CKD stage — CKD-EPI 2021
🙋
My check-up report says "eGFR 65". What is this number? I gather it has something to do with the kidney's filtration capacity.
🎓
Exactly. eGFR stands for estimated Glomerular Filtration Rate — how many millilitres of blood your glomeruli (the capillary tufts in the kidney) filter per minute, normalized to a body surface area of 1.73 m². A healthy young adult is around 100, below 80 is mild decline, and below 60 sustained for 3 months is the definition of chronic kidney disease (CKD). Measuring it directly needs a 24-hour creatinine clearance test, which is awkward, so the CKD-EPI equation estimates it from serum creatinine, age and sex — and that has become the world standard.
🙋
I hear serum creatinine is heavily affected by muscle mass. So how can it be a kidney marker?
🎓
Good point. Creatinine is a by-product of muscle creatine-phosphate metabolism — it is released into the blood at a roughly constant rate and filtered ~100% by the glomeruli. So when kidney function drops, it accumulates. But people with more muscle produce more, so a bodybuilder can have a high Scr with perfectly healthy kidneys. That is why CKD-EPI corrects with age and sex. For very atypical body composition, a cystatin C–based eGFR-Cys, which is independent of muscle mass, is recommended.
🙋
When I switch the formula between "CKD-EPI 2009" and "2021", the race selector changes whether it does anything. What changed?
🎓
That switch is a major medical-history event. The CKD-EPI 2021 equation, published in NEJM in 2021, removed the race factor entirely (the 1.159× multiplier the older equation applied for Black patients). The reasoning is that race is a social construct without firm biological grounds for altering kidney function. In practice, eGFR comes out a little lower for Black patients, and some who were G2 under the old equation are now G3a, with downstream effects on drug dosing and transplant listing.
🙋
UACR also shows up. Why does the tool flag "high risk" when eGFR is good but UACR is bad?
🎓
Modern prognosis uses the KDIGO heat map — a two-axis grid of eGFR (G1-G5) and UACR (A1-A3). At the same eGFR of 60, A1 (UACR <30) is low risk while A3 (≥300, overt albuminuria) carries several-fold higher dialysis and cardiovascular mortality rates. Diabetic nephropathy in particular tends to leak albumin before eGFR drops, which is why UACR is the cornerstone of early detection. SGLT2 inhibitors (dapagliflozin, empagliflozin) have been shown to lower UACR by 30-50% in DAPA-CKD and EMPA-KIDNEY, and they are now the workhorse of CKD therapy.
🙋
Once eGFR drops below 15, does dialysis automatically begin? Or can it be delayed?
🎓
eGFR <15 is KDIGO G5 (ESRD — End-Stage Renal Disease) and is the prep phase for dialysis, but the actual start is decided individually based on symptoms (uremia, fluid overload, hyperkalemia, metabolic acidosis). Some patients tolerate eGFR 5-7 on conservative therapy; others start at 10 because symptoms are severe. The US adds ~130,000 new dialysis patients per year at a $90B cost, and the goal of conservative therapy (ACEi/ARB, SGLT2 inhibitors, phosphate management with lanthanum, erythropoiesis-stimulating agents like Epogen) is to delay initiation by 2-5 years. Sysmex and Beckman benchtop creatinine analyzers are the clinical workhorses.
Frequently Asked Questions
The 2009 equation included a race factor (1.159 multiplier for Black patients). The 2021 version (Inker et al., NEJM) removed the race coefficient entirely, motivated by the position that race is a social construct without firm biological basis for changing kidney function. The 2021 equation produces slightly lower eGFR values; some Black patients previously classified as G2 are now G3a, which affects drug dosing and transplant listing. The US NKF/ASN task force recommends immediate adoption of the 2021 equation.
Cockcroft-Gault (1976) estimates creatinine clearance (CrCl, mL/min) and uses body weight directly. The FDA drug label adjustments for renal function are still written predominantly in CrCl terms, so it remains the first choice for dosing DOACs (direct oral anticoagulants), vancomycin and gadolinium contrast agents. CKD-EPI eGFR is normalized to body surface area (mL/min/1.73m²) and is less suitable for drug dosing in patients with extreme body habitus. Modern practice uses each equation for the right context.
The KDIGO 2012 heat map stratifies prognosis on two axes: eGFR (G1-G5) and UACR (A1-A3). With the same eGFR of 60, UACR <30 (A1) is low risk while UACR ≥300 (A3) becomes high risk — dialysis and cardiovascular mortality rates jump several-fold. In diabetic nephropathy, albuminuria often appears before any eGFR decline, so UACR is essential for early screening. ACEi/ARB and SGLT2 inhibitors have been shown in multiple RCTs (DAPA-CKD, EMPA-KIDNEY) to reduce UACR by 30-50%.
eGFR <15 mL/min/1.73m² is KDIGO stage G5 (ESRD — End-Stage Renal Disease) and is the preparation phase for dialysis, but the actual start time is individualized by symptoms (uremia, fluid overload, hyperkalemia, metabolic acidosis). Some patients hold off until eGFR 5-7 on conservative therapy; others start at 10 because symptoms are severe. The US adds about 130,000 new dialysis patients per year at a cost near $90B, and modern practice tries to delay initiation by 2-5 years with ACEi/ARB, SGLT2 inhibitors, phosphate management and erythropoietin.
Real-World Applications
Nephrology outpatient practice: CKD is monitored 1-4 times a year using the two-axis combination of eGFR and UACR. From G3a onward, ACEi/ARB and SGLT2 inhibitors (dapagliflozin, empagliflozin) are added to slow progression. From G4 onward patients are referred to a nephrologist for planning dialysis initiation or transplantation, including vascular access creation. Benchtop creatinine analyzers from Sysmex, Beckman Coulter, Roche and Abbott are the clinical workhorses, with IDMS (Isotope Dilution Mass Spectrometry) standardization rolling out across Jaffe and enzymatic assays.
Drug dosing adjustment: Many drugs are renally cleared, so dosing must be adjusted to eGFR or CrCl. Vancomycin, gadolinium MRI contrast, metformin, DOACs (rivaroxaban, apixaban) and chemotherapy agents (cisplatin) are particularly strict, and FDA labels still specify thresholds in terms of Cockcroft-Gault CrCl. CKD-EPI eGFR is BSA-normalized, which makes it unsuitable for dosing in extreme body habitus, so clinical pharmacists check both equations.
Health policy for transplant and dialysis: The US has roughly 550,000 dialysis patients at an annual cost of $90B (about 7% of total Medicare spending), with ~130,000 new starts each year. Japan has 340,000 dialysis patients — among the highest per-capita rates in the world. UNOS transplant registration uses eGFR <20 as the listing threshold, with priority points calculated from eGFR and waiting time. Adoption of CKD-EPI 2021 has been shown to bring forward transplant listing for Black patients, with major implications for health equity.
Patient enrolment in pharma trials: Phase III trials for new CKD drugs (finerenone Kerendia, SGLT2 inhibitors, the endothelin-receptor antagonist sparsentan) stratify patients by eGFR 25-75 and UACR 200-5000. DAPA-CKD (NEJM 2020) and EMPA-KIDNEY (NEJM 2023) were landmark trials showing 30-40% reductions in dialysis progression with SGLT2 inhibitors, rewriting the CKD treatment paradigm. Being able to instantly check whether a given patient meets a trial's enrolment range is practically important.
Common Misconceptions and Pitfalls
The most frequent error is mechanically diagnosing CKD as soon as eGFR drops below 60. KDIGO requires "eGFR <60 or a urinary abnormality (albuminuria, haematuria, imaging abnormality) lasting 3 months or longer". Transient dehydration, intense exercise (raising creatinine production), and certain drugs (trimethoprim, cimetidine and corticosteroids that block creatinine secretion and falsely lower eGFR) all move the number around. A repeat measurement after 2-3 months is mandatory. Telling someone "you have CKD" after a single screening value of 58 is overdiagnosis.
Next, looking only at the absolute eGFR and ignoring its rate of change. In CKD the truly dangerous signal is the annual decline rate. Healthy aging gives 0.5-1.0 mL/min/1.73m² per year; a "rapid decliner" losing more than 5 mL/min/year is at sharply elevated short-term dialysis risk. An eGFR of 45 today is acceptable as stable G3a — but if it was 65 three months ago, that demands an acute work-up. The years-to-ESRD figure in this tool assumes a typical decline of 1.0 mL/min/year and must be replaced by each patient's measured slope in practice.
Finally, treating Cockcroft-Gault and CKD-EPI as interchangeable. Cockcroft-Gault returns CrCl in mL/min (not BSA-normalized) using body weight, and it overestimates in obesity and underestimates in lean elderly patients. CKD-EPI returns mL/min/1.73m² and is the standard for CKD staging, but using it for drug dosing in extreme body sizes causes misdosing. Clinical convention is: "CKD diagnosis → CKD-EPI", "drug dosing → Cockcroft-Gault CrCl", "atypical muscle mass → cystatin C–based eGFR-Cys". Switch between the four equations in this tool to feel the differences first-hand.
How to Use
Enter serum creatinine in mg/dL (typical range 0.6–1.2 for males, 0.5–1.1 for females)
Input patient age in years and select biological sex to apply appropriate equation coefficients
Enter body weight in kg if using Cockcroft-Gault; optional for MDRD and CKD-EPI 2009/2021
Input urine albumin-to-creatinine ratio (UACR) in mg/g to classify albuminuria stage
Review eGFR values across four simultaneous equations and corresponding CKD stage (1–5)
Check dialysis threshold flag when eGFR drops below 15 mL/min/1.73m²
Worked Example
72-year-old female, serum creatinine 1.8 mg/dL, weight 62 kg, UACR 145 mg/g. CKD-EPI 2021 yields eGFR 32 mL/min/1.73m² (Stage 3b CKD). MDRD equation estimates 28 mL/min/1.73m². Cockcroft-Gault calculates 24 mL/min/min without body surface area adjustment. Albuminuria category: moderate (30–300 mg/g UACR). High-risk profile: yes. Dialysis not yet needed; estimated 8–12 years to ESRD at current decline rate of 2–3 mL/min/1.73m² annually if untreated.
Practical Notes
CKD-EPI 2021 equation includes race-free coefficient; preferred for new diagnoses in nephrology and primary care
Cockcroft-Gault overestimates eGFR in elderly and obese patients; use adjusted body weight when BMI exceeds 30
UACR greater than 30 mg/g indicates albuminuria risk; ACE inhibitor or ARB initiation recommended regardless of eGFR stage
Serial creatinine trends over 3–6 months improve CKD progression prediction more than single snapshot values
Muscle mass variation (cachexia, amputees, athletes) can shift creatinine-based eGFR by 10–20%; cystatin C equation valuable for validation