MDRD GFR Calculator

How to Use This Calculator

1. Enter serum creatinine (mg/dL), age, sex, and race for MDRD-4.
2. Switch to MDRD-6 tab and add BUN and albumin for the 6-variable estimate.
3. Use "vs CKD-EPI" tab to compare MDRD-4 against CKD-EPI 2021.
4. Professional tier shows all three equations side-by-side with KDIGO staging and drug dosing notes.

Formula

Example

Frequently Asked Questions

• What is the MDRD equation and how is it different from CKD-EPI? ▼
  The MDRD (Modification of Diet in Renal Disease) equation is a regression formula developed to estimate glomerular filtration rate (GFR) from serum creatinine, age, sex, and race without requiring a urine collection. The most widely used form, MDRD-4 (Levey et al. 2006), is: GFR = 175 × Cr^−1.154 × Age^−0.203 × 0.742 (if female) × 1.212 (if Black). The original MDRD-6 (Levey et al. 1999) additionally incorporated BUN and albumin for slightly improved precision. CKD-EPI 2021 (Inker et al. NEJM 2021) is a newer equation using the same four variables but a spline-based formula that more accurately estimates GFR across the full range, especially at GFR >60 mL/min/1.73m². Key differences: MDRD systematically underestimates GFR at values above 60, causing overclassification of CKD stage G1 and G2. CKD-EPI has better bias and precision in people with higher GFR and was updated in 2021 to remove the race coefficient entirely, addressing equity concerns. For these reasons, KDIGO 2024 and NKF/ASN 2021 now recommend CKD-EPI 2021 as the preferred eGFR equation for clinical use, though MDRD remains embedded in many drug dosing references.
• Why was the MDRD equation developed? ▼
  The MDRD equation emerged from the Modification of Diet in Renal Disease study, a multicenter randomized controlled trial funded by the NIDDK and published in 1994. The trial's primary aim was to determine whether protein restriction or strict blood pressure control slowed CKD progression. A secondary outcome was the development of a practical GFR estimation formula that eliminated the need for timed urine collections (24-hour creatinine clearance), which are cumbersome, prone to collection error, and impractical in clinical practice. Before MDRD, Cockcroft-Gault (1976) was the standard, but it estimates creatinine clearance (not GFR) and is not indexed to body surface area. The original MDRD study enrolled 1,628 patients with CKD (mean GFR ~40 mL/min/1.73m²), and the resulting equations were validated against measured GFR by iothalamate clearance. The MDRD-4 simplification (Levey et al. 2006) retained only four variables (removing BUN and albumin) with minimal loss of accuracy, making it suitable for automated laboratory reporting. From the mid-2000s, MDRD-4 became the default eGFR equation automatically reported with serum creatinine results in the US, UK, Australia, and most of Europe, dramatically improving CKD detection and staging in clinical practice.
• Should I still use MDRD or switch to CKD-EPI? ▼
  For clinical decision-making, CKD-EPI 2021 should be the default equation. The 2021 NKF/ASN joint task force and the KDIGO 2024 CKD Clinical Practice Guidelines both explicitly recommend CKD-EPI 2021 (race-free version) as the preferred equation for estimating GFR in adults. CKD-EPI performs better than MDRD across the entire GFR range, particularly at GFR >60 mL/min/1.73m² where MDRD substantially underestimates kidney function — leading to unnecessary CKD diagnoses and inappropriate drug dose reductions. In practice, MDRD remains relevant in two main contexts: (1) Drug dosing: many pharmacokinetic studies, drug package inserts, and renal dosing references were developed using MDRD or Cockcroft-Gault, so clinicians should use the equation matched to the reference population for a specific medication. Abruptly switching formulas can cause systematic errors in drug dose selection. (2) Research and historical comparisons: studies published before ~2012 used MDRD as their eGFR, so understanding MDRD is essential for interpreting older literature. The practical approach: use CKD-EPI 2021 for diagnosis, staging, and most clinical decisions; consult drug-specific guidance for dosing; understand MDRD for interpreting historical data.
• Why does MDRD include race, and is the race coefficient valid? ▼
  The MDRD race coefficient (×1.212 for Black individuals) was derived empirically: in the original MDRD study population, Black participants had higher measured GFR for the same serum creatinine level compared to non-Black participants. The biological explanation invoked at the time was that Black individuals have higher average muscle mass and creatinine generation, causing creatinine-based GFR estimates to underestimate true GFR unless corrected upward. However, this race adjustment has been extensively criticized on scientific and equity grounds. First, race is a social construct, not a biological variable — individuals differ in muscle mass, creatinine excretion, and tubular creatinine secretion for many reasons unrelated to race. The coefficient was derived from a small, selected population and applied categorically across a heterogeneous group. Second, in practice, applying the Black race multiplier raises eGFR estimates, which can delay CKD diagnosis, delay referral to nephrology, and disadvantage Black patients in kidney transplant allocation (which uses eGFR thresholds). A 2021 NKF/ASN task force concluded that no race or ethnicity variable should be included in GFR-estimating equations. CKD-EPI 2021 achieves similar or better performance without a race term. MDRD's race coefficient should therefore be used only for historical context.
• When is the MDRD-6 equation preferred over MDRD-4? ▼
  The MDRD-6 equation adds BUN and serum albumin to the four variables of MDRD-4. In the original Levey et al. 1999 derivation study, MDRD-6 had slightly lower bias and better precision than MDRD-4, particularly in specific clinical subgroups. There are two scenarios where MDRD-6 may add value. First, patients with significant malnutrition or low muscle mass (low albumin, low BUN): these conditions cause serum creatinine to underestimate true kidney function (creatinine is lower than expected for the actual GFR). MDRD-6's albumin and BUN terms partially correct for this. Second, patients with markedly abnormal protein metabolism (cirrhosis, severe cachexia, or high protein intake): BUN reflects protein catabolism and dietary protein intake, so including it can refine the estimate. In clinical practice, however, MDRD-6 is rarely reported by laboratories because it requires two additional values (BUN and albumin), increasing complexity without providing dramatic accuracy gains over MDRD-4. Most laboratories auto-report MDRD-4 or CKD-EPI. MDRD-6 is primarily of academic interest and is available in research calculators like MDCalc. If BUN and albumin are available, CKD-EPI with cystatin C (when cystatin C is measured) outperforms MDRD-6 in most contemporary studies and should be preferred.

Related Calculators