eGFR equations: CKD-EPI 2021 vs MDRD vs Cockcroft-Gault

For an 85-year-old woman of 60 kg and 160 cm with creatinine 0.8 mg/dL, CKD-EPI 2021 gives 72 mL/min/1.73 m² and MDRD 68. Cockcroft-Gault gives 49 mL/min, 29% below CKD-EPI de-indexed to her 1.63 m² (68 mL/min). At 30 the same woman gets 102, 84 and 97: Cockcroft-Gault falls furthest with age.

Line chart of estimated kidney function against age from 20 to 90 years, for a woman of 60 kg and 160 cm with creatinine 1.0 mg/dL. Cockcroft-Gault falls fastest, from 85 to 35 mL/min. CKD-EPI 2021 falls from 83 to 54 mL/min/1.73 m² and MDRD from 71 to 52 mL/min/1.73 m². Grey dashed reference lines at 60, 45 and 30 mark where KDIGO categories G3a (45 to 59), G3b (30 to 44) and G4 (15 to 29) begin. — CKD-EPI 2021 and MDRD are per 1.73 m² of body surface; Cockcroft-Gault is a clearance in mL/min for this woman’s 60 kg. Every plotted point is a row of the table. Image © ScienceQuest, CC BY 4.0: free to reuse with credit and a link to this page, as the image licence explains.

By age, at creatinine 1.0 mg/dL

The three estimates and the KDIGO category each gives, for a woman of 60 kg and 160 cm with creatinine 1.0 mg/dL (88 µmol/L), every 10 years from 20 to 90. The chart plots the three estimates.
Age (years)	Estimate			KDIGO GFR category
Age (years)	CKD-EPI 2021 mL/min/1.73 m²	MDRD mL/min/1.73 m²	Cockcroft-Gault mL/min	CKD-EPI	MDRD	C-G*
20	83	71†	85	G2	G2	G2
30	78	65†	78	G2	G2	G2
40	73	61†	71	G2	G2	G2
50	69	59	64	G2	G3a	G2
60	64	57	57	G2	G3a	G3a
70	61	55	50	G2	G3a	G3a
80	57	53	43	G3a	G3a	G3b
90	54	52	35	G3a	G3a	G3b

† 60 or above, where NKDEP advised laboratories to report no MDRD number. * Cockcroft-Gault is a clearance in mL/min, not a GFR per 1.73 m², so KDIGO did not write its categories for it; the column shows the band you get by reading it as one.

Women, 60 kg and 160 cm

CKD-EPI 2021 and MDRD eGFR in mL/min/1.73 m² and Cockcroft-Gault (C-G) creatinine clearance in mL/min, for a woman of 60 kg and 160 cm, by serum creatinine and age.
Creatinine, mg/dL (µmol/L)	Age 30			Age 50			Age 70			Age 85
Creatinine, mg/dL (µmol/L)	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G
0.5 (44)	129	145†	156	114	131†	128	101	122†	99	92	117†	78
0.6 (53)	124	117†	130	109	106†	106	97	99†	83	88	95†	65
0.7 (62)	119	98†	111	105	89†	91	93	83†	71	85	80†	56
0.8 (71)	102	84†	97	90	76†	80	79	71†	62	72	68†	49
0.9 (80)	88	74†	87	78	66†	71	69	62†	55	63	60†	43
1.0 (88)	78	65†	78	69	59	64	61	55	50	55	53	39
1.2 (106)	62	53	65	55	48	53	49	44	41	44	43	32
1.5 (133)	48	41	52	42	37	43	37	34	33	34	33	26
2.0 (177)	34	29	39	30	26	32	26	25	25	24	24	19
3.0 (265)	21	18	26	18	17	21	16	15	17	15	15	13

† 60 or above, where NKDEP advised laboratories to report no MDRD number. µmol/L is mg/dL × 88.4, rounded to the whole number.

Men, 80 kg and 178 cm

CKD-EPI 2021 and MDRD eGFR in mL/min/1.73 m² and Cockcroft-Gault (C-G) creatinine clearance in mL/min, for a man of 80 kg and 178 cm, by serum creatinine and age.
Creatinine, mg/dL (µmol/L)	Age 30			Age 50			Age 70			Age 85
Creatinine, mg/dL (µmol/L)	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G	CKD-EPI	MDRD	C-G
0.5 (44)	141	195†	244	124	176†	200	110	164†	156	100	158†	122
0.6 (53)	133	158†	204	118	143†	167	104	133†	130	95	128†	102
0.7 (62)	127	132†	175	112	119†	143	99	111†	111	90	107†	87
0.8 (71)	122	114†	153	108	102†	125	95	96†	97	87	92†	76
0.9 (80)	118	99†	136	104	89†	111	92	83†	86	84	80†	68
1.0 (88)	104	88†	122	92	79†	100	81	74†	78	74	71†	61
1.2 (106)	83	71†	102	74	64†	83	65	60†	65	59	58	51
1.5 (133)	64	55	81	56	50	67	50	46	52	45	44	41
2.0 (177)	45	39	61	40	36	50	35	33	39	32	32	31
3.0 (265)	28	25	41	25	22	33	22	21	26	20	20	20

† 60 or above, where NKDEP advised laboratories to report no MDRD number. µmol/L is mg/dL × 88.4, rounded to the whole number.

By body weight, at 70 and creatinine 1.0 mg/dL

A woman of 160 cm and a man of 178 cm, both aged 70 with creatinine 1.0 mg/dL, by body weight: body surface area, CKD-EPI 2021 per 1.73 m², the same de-indexed to the person’s own body surface area, and Cockcroft-Gault.
Weight (kg)	Woman, 160 cm				Man, 178 cm
Weight (kg)	BSA m²	CKD-EPI 2021 mL/min/1.73 m²	CKD-EPI de-indexed mL/min	Cockcroft-Gault mL/min	BSA m²	CKD-EPI 2021 mL/min/1.73 m²	CKD-EPI de-indexed mL/min	Cockcroft-Gault mL/min
50	1.49	61	52	41	1.57	81	74	49
70	1.76	61	62	58	1.86	81	87	68
90	2.00	61	70	74	2.11	81	99	88
120	2.31	61	81	99	2.44	81	114	117

De-indexed: CKD-EPI × BSA ÷ 1.73, with BSA by Mosteller. That is the figure to set beside Cockcroft-Gault.

Why the formulas disagree

CKD-EPI 2021 and MDRD estimate glomerular filtration rate per 1.73 m² of body surface, the average of 25-year-old Americans in 1927, so neither takes weight. Cockcroft-Gault estimates creatinine clearance in mL/min from actual body weight. At age 70 and creatinine 1.0 mg/dL, the 160 cm woman’s CKD-EPI stays at 61 mL/min/1.73 m² from 50 to 120 kg, while her Cockcroft-Gault climbs from 41 to 99 mL/min. To set the two side by side, multiply CKD-EPI by her body surface area over 1.73: 81 mL/min at 120 kg.

Cockcroft-Gault subtracts age from 140, so the chart’s woman loses 58% of her clearance between 20 and 90. CKD-EPI multiplies by 0.9938 for each year and loses 35%; MDRD takes age to the power −0.203 and loses 26%. Her Cockcroft-Gault falls below CKD-EPI de-indexed to her 1.63 m² at 47. On the chart her Cockcroft-Gault and CKD-EPI lines cross at 31, because the chart sets mL/min against mL/min/1.73 m².

MDRD gives creatinine one exponent, and it reads low at higher GFR: the MDRD Study developed and validated it in 1,628 people with chronic kidney disease whose measured GFR averaged 39.8 mL/min/1.73 m². CKD-EPI bends at 0.7 mg/dL for women and 0.9 for men. For the 50-year-old woman at 0.7 mg/dL, CKD-EPI gives 105 mL/min/1.73 m² and MDRD 89; at 3.0 mg/dL they give 18 and 17. Cockcroft-Gault predates standardised creatinine assays, and KDIGO notes that nobody can re-express it for them.

The formulas

CKD-EPI 2021 [1]: $\mathrm{eGFR} = 142 \times \min\!\left(\tfrac{S_{Cr}}{\kappa}, 1\right)^{\alpha} \times \max\!\left(\tfrac{S_{Cr}}{\kappa}, 1\right)^{-1.200} \times 0.9938^{\,\text{age}} \times 1.012\ [\text{female}]$
MDRD, re-expressed for IDMS creatinine [2]: $\mathrm{eGFR} = 175 \times S_{Cr}^{-1.154} \times \text{age}^{-0.203} \times 0.742\ [\text{female}]$
Cockcroft-Gault [4]: $C_{Cr} = \frac{(140 - \text{age}) \times \text{weight}}{72 \times S_{Cr}} \times 0.85\ [\text{female}]$
De-indexing, with BSA by Mosteller [5]: $\mathrm{GFR}_{\text{mL/min}} = \mathrm{eGFR} \times \frac{\mathrm{BSA}}{1.73},$ $\mathrm{BSA} = \sqrt{\frac{\text{height} \times \text{weight}}{3600}}$

SCr is serum creatinine in mg/dL, age in years, weight in kg, height in cm. CKD-EPI and MDRD give mL/min/1.73 m², Cockcroft-Gault mL/min.

Where each formula holds

Each formula’s sample, the people it holds for, and where it does not apply
Formula	Fitted in	Holds for	Not for
CKD-EPI 2021	8,254 adults in 10 studies, 31.5% of them Black, refitted without race; validated in 4,050 adults in 12 other studies	Adults 18 and over with a stable creatinine on an IDMS-traceable assay; NKF and ASN advised every US laboratory to adopt it	Acute kidney injury and the other limits KDIGO 2012 Table 11 sets for any creatinine estimate: extremes of muscle mass or body size, high-protein diets and creatine supplements, drugs such as trimethoprim and cimetidine. Least certain over 70: the 2009 data it refits had relatively few participants that old (Levey 2009)
MDRD (175, IDMS)	Developed and validated in 1,628 people with chronic kidney disease in the MDRD Study, measured GFR averaging 39.8 mL/min/1.73 m²	Adults with chronic kidney disease; its 2006 paper found it reasonably accurate where measured GFR was below 90	Acute kidney injury and the other KDIGO Table 11 limits, as for CKD-EPI; a figure of 60 or above, which NKDEP advised laboratories not to report as a number; higher GFR, where it reads low
Cockcroft-Gault	Creatinine excretion in 249 patients aged 18 to 92, for adult men, with 15% less for women; checked against measured clearance in 236	Adults, as a creatinine clearance in mL/min for the person’s own weight, not adjusted to 1.73 m²	Acute kidney injury and the other KDIGO Table 11 limits, extremes of body size among them; standardised (IDMS) creatinine, which it predates; anything read as a GFR per 1.73 m²

Calculators and simulators

A study tool, not a clinical one

This page exists for understanding what an estimated GFR is, why three equations disagree for the same patient, and where each one stops being reliable. It is not clinical decision support and must not be used to treat anybody.

A row in these tables is not your patient. Each row fixes the other inputs at the values its caption gives, and a real patient differs in all of them at once, so work a patient’s own figures in the calculator.
Check the units before you trust a number. American and Indian reports mostly use mg/dL, while British, Canadian and Australian reports use mmol/L or µmol/L, and a unit mismatch is the most common way an otherwise correct formula produces a dangerous answer.

Three more cautions

Each formula has a population. The table of where each one holds gives the people each was fitted in, and each is least reliable outside them.
Know which equation your course or laboratory uses. The ones compared here give different numbers for the same patient, which is what the tables show.
Every cell is worked by the calculators’ own code. If a figure does not match what you were taught, check it against the formulas above, and tell us if we have it wrong. How the tools are checked is set out in the editorial policy.

Sources

Inker LA, Eneanya ND, Coresh J, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021;385(19):1737-1749. doi:10.1056/NEJMoa2102953
Levey AS, Coresh J, Greene T, et al. Using standardized serum creatinine values in the Modification of Diet in Renal Disease Study equation for estimating glomerular filtration rate. Ann Intern Med. 2006;145(4):247-254. doi:10.7326/0003-4819-145-4-200608150-00004
Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-612. The 2009 CKD-EPI equation, which the 2021 equation refits without race. doi:10.7326/0003-4819-150-9-200905050-00006
Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31-41. doi:10.1159/000180580
Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987;317(17):1098. The letter was not accessible to us; the next source prints its formula. doi:10.1056/NEJM198710223171717
Mohamed AAA, van Londen M, Stevens J, Heerspink HJL, Gansevoort RT. Indexing for body surface area when assessing kidney function. Nephrol Dial Transplant. 2026;41(9):1595-1605. Table 1 gives Mosteller’s formula as (weight^0.5 × height^0.5)/60, the square root above. doi:10.1093/ndt/gfag063
KDIGO CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl. 2013;3(1):1-150. The G1 to G5 categories, the rounding and reporting advice, and the limits of creatinine estimates. https://kdigo.org/wp-content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf
Delgado C, Baweja M, Crews DC, et al. A unifying approach for GFR estimation: recommendations of the NKF-ASN Task Force on reassessing the inclusion of race in diagnosing kidney disease. Am J Kidney Dis. 2022;79(2):268-288. doi:10.1053/j.ajkd.2021.08.003