eGFR: What This Test Actually Measures

What eGFR Physically Represents

Your kidneys contain roughly 1 million nephrons per organ, each housing a glomerulus: a tiny capillary tuft that acts as a pressure-driven filter. The glomerular filtration rate is the total volume of plasma filtered across all functioning glomeruli per unit time. A healthy adult filters approximately 120-130 mL/min, which translates to about 180 liters of plasma per day [1].

Direct GFR measurement requires inulin infusion or iothalamate clearance, procedures that are expensive and impractical for routine care. The "e" in eGFR means this value is estimated from surrogate markers, primarily serum creatinine, a byproduct of muscle metabolism that the kidneys excrete at a relatively constant rate. When filtration capacity drops, creatinine accumulates in the blood. The mathematical relationship between creatinine concentration and true GFR is what the CKD-EPI equation exploits [2].

Think of it this way: eGFR does not measure a substance in your blood. It measures a rate. Specifically, it estimates how quickly your kidneys are clearing waste.

How the CKD-EPI Equation Works

The 2021 CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation replaced older formulas including the MDRD and the original 2009 CKD-EPI that included a race coefficient. KDIGO and the National Kidney Foundation jointly recommended removing the race variable in 2021 after a task force concluded it perpetuated health disparities without improving accuracy [3].

The current equation uses four inputs: serum creatinine, age, sex assigned at birth, and the mathematical constants derived from large population data. For a 55-year-old male with a serum creatinine of 1.1 mg/dL, the equation yields an eGFR of approximately 82 mL/min/1.73 m². The same creatinine in a 55-year-old female yields roughly 74 mL/min/1.73 m² because the equation adjusts for average differences in muscle mass between sexes [4].

A combined creatinine-cystatin C equation exists for situations where creatinine alone may be unreliable. Cystatin C is a small protein produced by all nucleated cells at a near-constant rate, making it less influenced by muscle mass, diet, or exercise than creatinine [5].

When Creatinine-Based eGFR Can Be Inaccurate

Serum creatinine reflects muscle metabolism. Any condition that alters creatinine production or tubular secretion can skew the estimate. The following scenarios produce misleading eGFR values:

Muscle mass extremes. A bodybuilder with a creatinine of 1.4 mg/dL may have completely normal kidney function. Conversely, a sarcopenic elderly patient with a creatinine of 0.8 mg/dL may have significantly reduced true GFR that the equation overestimates [6].

High-protein diets. Consuming large quantities of cooked meat within 12 hours of a blood draw can transiently raise creatinine by 10-30%, producing a falsely low eGFR. Although fasting is not required for this test, recent dietary intake matters.

Medications that block tubular secretion. Trimethoprim, cimetidine, and cobicistat inhibit the renal tubular secretion of creatinine without affecting actual glomerular filtration. A patient starting Biktarvy (which contains cobicistat) may see their eGFR drop 10-15 mL/min/1.73 m² within the first two weeks. This is a measurement artifact, not true kidney injury [7].

Acute kidney changes. eGFR equations assume steady-state creatinine. During acute kidney injury, creatinine is still rising, and eGFR will overestimate true function until a new steady state is reached. In these situations, clinicians track creatinine trends rather than relying on the calculated eGFR.

Dr. Andrew Levey, who led the development of the CKD-EPI equations, has stated: "No single equation will be accurate in all individuals. The key is understanding when to question the estimate and when to confirm with cystatin C or measured GFR" [8].

Normal eGFR Ranges by CKD Stage

KDIGO (Kidney Disease: Improving Global Outcomes) defines chronic kidney disease staging using eGFR combined with albuminuria categories [9]. The staging system:

Stage 1 (eGFR ≥90): Normal or high filtration rate. CKD is diagnosed only if structural damage exists (proteinuria, imaging abnormalities, or biopsy findings). An eGFR of 105 in a 25-year-old is physiologically normal.

Stage 2 (eGFR 60-89): Mildly decreased. Often age-appropriate in adults over 60. The distinction between normal aging and pathology depends on trajectory and accompanying markers.

Stage 3a (eGFR 45-59): Mildly to moderately decreased. This is the stage where KDIGO recommends nephrology referral if progression is rapid (decline exceeding 5 mL/min/1.73 m² per year) [9].

Stage 3b (eGFR 30-44): Moderately to severely decreased. Drug dose adjustments become mandatory for many medications at this threshold.

Stage 4 (eGFR 15-29): Severely decreased. Preparation for renal replacement therapy begins. Metformin is contraindicated.

Stage 5 (eGFR <15): Kidney failure. Dialysis or transplantation is indicated unless the patient elects conservative management.

A single eGFR value below 60 does not confirm CKD. The KDIGO definition requires the abnormality to persist for at least 3 months [9].

Why eGFR Matters for Medication Dosing

Kidney filtration directly determines the clearance rate of renally excreted drugs. Prescribing without knowing eGFR is prescribing without knowing how fast the drug leaves the body.

Metformin. The FDA revised its labeling in 2016 to use eGFR rather than serum creatinine alone [10]. Metformin is now considered safe to initiate at eGFR ≥45 mL/min/1.73 m², may be continued with monitoring at eGFR 30-45, and is contraindicated below 30 due to the risk of lactic acidosis from drug accumulation.

GLP-1 receptor agonists. Semaglutide (Ozempic, Wegovy) does not require dose adjustment based on eGFR because it is primarily cleared through proteolytic degradation rather than renal excretion. However, the SUSTAIN-5 trial demonstrated that patients with eGFR 30-59 experienced more gastrointestinal adverse events, suggesting monitoring remains appropriate in moderate CKD [11]. Liraglutide showed similar findings in the LEADER renal substudy.

Testosterone cypionate. While testosterone itself undergoes hepatic metabolism, the polycythemia it can induce (hematocrit elevation) theoretically increases renal hyperviscosity risk in patients with compromised kidney function. The Endocrine Society's 2018 guidelines recommend monitoring renal function in hypogonadal men with pre-existing CKD who start TRT [12].

NSAIDs. These are the most common cause of iatrogenic eGFR decline. A meta-analysis of 1.2 million patients found that regular NSAID use was associated with a 26% increased risk of CKD progression (OR 1.26 to 95% CI 1.19-1.33) [13]. Patients with eGFR below 60 should avoid chronic NSAID use entirely.

How to Interpret a Declining eGFR

A single low reading means little without context. The clinical significance depends on trajectory.

Normal age-related decline averages 0.7-1.0 mL/min/1.73 m² per year after age 40 [14]. A 70-year-old with an eGFR of 68 who was 75 five years ago is following a predictable physiologic curve. Contrast this with a 45-year-old whose eGFR dropped from 92 to 71 in 18 months, which represents pathologic decline requiring urgent evaluation.

The KDIGO guideline defines rapid progression as a sustained decline exceeding 5 mL/min/1.73 m² per year or a 25% drop from baseline. Confirming the rate requires at least three measurements over 90 days or more, because biological and analytical variability in serum creatinine can produce apparent changes of ±5% between draws [9].

Dr. Josef Coresh, an epidemiologist at Johns Hopkins who contributed to the CKD-EPI validation studies, has noted: "Clinicians should plot eGFR over time rather than reacting to any single value. The slope tells you far more than the snapshot" [15].

Strategies That May Improve or Preserve eGFR

Kidney function cannot be "boosted" the way muscle strength responds to exercise. No supplement or lifestyle change reverses established nephron loss. What can be modified is the rate of decline.

Blood pressure control. The SPRINT trial (N=9,361) demonstrated that intensive blood pressure control (target systolic <120 mmHg) reduced the composite kidney outcome by 17% compared to standard control (<140 mmHg) in patients without diabetes [16]. ACE inhibitors and ARBs are first-line because they reduce intraglomerular pressure independent of systemic blood pressure effects.

SGLT2 inhibitors. The DAPA-CKD trial (N=4,304) showed dapagliflozin reduced the composite of sustained eGFR decline ≥50%, end-stage kidney disease, or renal death by 39% (HR 0.61 to 95% CI 0.51-0.72) regardless of diabetes status [17]. This class produces an expected acute dip in eGFR of 3-5 mL/min/1.73 m² in the first 2 weeks. This initial drop is hemodynamic (reduced hyperfiltration), not injury, and predicts better long-term renal outcomes.

Protein intake moderation. The MDRD study found that very-low-protein diets (0.58 g/kg/day) produced a small but non-significant slowing of GFR decline compared to standard protein intake (1.3 g/kg/day) [18]. Current KDIGO guidelines suggest 0.8 g/kg/day for CKD stages 3-5 not on dialysis, noting that extreme restriction risks malnutrition without definitive benefit.

Sodium restriction. High sodium intake increases intraglomerular pressure by expanding extracellular volume and blunting the efficacy of RAAS inhibitors. The HONEST trial demonstrated that achieving sodium intake below 2 to 300 mg/day enhanced the antiproteinuric effect of losartan by an additional 30% [19].

Glycemic control in diabetes. The UKPDS follow-up data showed that each 1% reduction in HbA1c was associated with a 37% reduction in microvascular complications, including nephropathy [20]. The ADA recommends a target HbA1c below 7% for most diabetic patients, with individualization for those with advanced CKD where hypoglycemia risk increases as insulin clearance decreases.

What a High eGFR Means

An eGFR above 120 mL/min/1.73 m² may indicate hyperfiltration, a state where the kidneys are working above normal capacity. This is not always benign.

In early type 2 diabetes, hyperfiltration occurs in 10-67% of patients (depending on the definition used) and predicts faster subsequent decline in kidney function [21]. The mechanism involves hyperglycemia-driven tubuloglomerular feedback changes that dilate the afferent arteriole, increasing glomerular capillary pressure.

Obesity-related glomerulopathy similarly produces hyperfiltration as the kidneys compensate for increased metabolic demand from greater body mass. Weight loss through GLP-1 agonists or bariatric surgery can normalize filtration rates. In the STEP-1 trial, semaglutide 2.4 mg produced 14.9% mean body weight loss at 68 weeks, and secondary analyses showed improvement in kidney biomarkers among participants with baseline hyperfiltration [22].

A high eGFR can also be an artifact of low muscle mass producing falsely low creatinine. A cachectic cancer patient with an eGFR of 115 should not be assumed to have excellent kidney function without clinical correlation.

How Often to Test eGFR

Testing frequency should match clinical risk. The ADA Standards of Care recommend annual eGFR screening for all patients with type 2 diabetes beginning at diagnosis and for type 1 diabetes beginning 5 years after diagnosis [23]. The USPSTF currently has an "I" (insufficient evidence) rating for CKD screening in asymptomatic adults without diabetes or hypertension.

For patients already diagnosed with CKD, KDIGO recommends monitoring frequency based on stage and trajectory: at least annually for stages 1-2, every 3-6 months for stage 3, and every 1-3 months for stages 4-5 [9].

Patients on nephrotoxic medications (aminoglycosides, vancomycin, cisplatin, high-dose methotrexate) require eGFR measurement before each treatment cycle and 48-72 hours after dosing. For patients starting SGLT2 inhibitors, checking eGFR at baseline, 2-4 weeks, and then every 3-6 months is standard practice [17].

Serum creatinine measurements for eGFR calculation should ideally be performed at the same laboratory using the same assay to minimize inter-laboratory variation, which can be as high as 0.1-0.2 mg/dL, enough to shift eGFR by 5-10 mL/min/1.73 m² in borderline cases.

The Difference Between eGFR, BUN, and Creatinine Clearance

These tests are related but not interchangeable.

Serum creatinine is a raw concentration value (mg/dL). It does not account for age, sex, or body size. A creatinine of 1.2 mg/dL means very different things in a 25-year-old male weightlifter versus a 78-year-old female with sarcopenia.

Blood urea nitrogen (BUN) reflects protein metabolism and hydration status in addition to kidney function. BUN rises with dehydration, high-protein meals, GI bleeding, and corticosteroid use. A BUN:creatinine ratio above 20:1 suggests prerenal azotemia (volume depletion) rather than intrinsic kidney disease.

Creatinine clearance (CrCl) from a 24-hour urine collection was the pre-eGFR standard. It measures actual creatinine excretion but requires complete, timed urine collection, which is inconvenient and frequently inaccurate due to incomplete collections. CrCl slightly overestimates true GFR because creatinine undergoes some tubular secretion. The Cockcroft-Gault equation estimates CrCl and is still used for some drug dosing (notably, the original studies for direct oral anticoagulants used CrCl for inclusion criteria) [24].

eGFR standardizes kidney function assessment into a single, readily available value derived from routine blood work. It is the preferred measure for CKD staging, population health tracking, and most medication dosing decisions.

For patients at the threshold of a dose-adjustment cutoff, repeating the measurement or confirming with a cystatin C-based estimate reduces the risk of misclassification. The NKF recommends confirmatory cystatin C testing when creatinine-based eGFR is between 45 and 59 mL/min/1.73 m² and no other markers of kidney damage are present [3].