eGFR and Creatinine: What the Numbers Can—and Cannot—Show

A serum creatinine result is a measured concentration in blood. eGFR— estimated glomerular filtration rate—is a calculated estimate of filtration, usually derived from creatinine plus other inputs. They are connected, but they are not the same test and neither number, by itself, diagnoses chronic kidney disease or precisely measures kidney function.

Read the pair as an evidence chain: what was measured, which equation produced the estimate, whether the marker was reliable in that context, what urine and other kidney evidence exists, and whether the finding is chronic or recent. Do not replace that chain with a universal creatinine range, a single eGFR cutoff, or an “optimal kidney number.”

Start by separating four different facts

A portal may place several kidney-related labels next to each other. Preserve them as distinct evidence:

• Serum creatinine is the laboratory measurement. The National Institute of Diabetes and Digestive and Kidney Diseases describes creatinine as a waste product from normal muscle breakdown that the kidneys remove from blood. The result is a concentration with a unit and report interval.
• eGFRcr is an estimate calculated from creatinine. For adults, the 2021 CKD-EPI creatinine equation uses age, sex, and standardized serum creatinine; it does not include a race coefficient.
• eGFRcys is an estimate based on cystatin C, another filtration marker. eGFRcr-cys combines creatinine and cystatin C. These labels reveal which marker or markers informed the estimate.
• Measured GFR uses the clearance of an administered filtration marker. It is not the routine serum creatinine measurement and should not be silently substituted for an eGFR label.

That naming distinction matters. “Creatinine,” “creatinine clearance,” “eGFRcr,” and “eGFRcr-cys” are not interchangeable result names. Keep the wording and unit from the source report instead of reducing all of them to one generic kidney-function row.

How serum creatinine becomes an eGFR estimate

The NIDDK's current guide to adult eGFR equations lists race-free CKD-EPI equations for people age 18 and older. The 2021 eGFRcr equation combines standardized serum creatinine with age and sex. The combined 2021 eGFRcr-cys equation also uses cystatin C.

The equation estimates filtration from how marker concentrations behave across studied populations. It does not turn creatinine into a direct filtration measurement. NIDDK explicitly cautions that eGFR is an estimate, not a precise measure, and that estimating equations become less accurate at higher GFR.

The 2021 race-free equations came from a large CKD-EPI collaboration. In the peer-reviewed development and validation study, the creatinine equation was developed using 8,254 participants, the combined creatinine–cystatin C equation using 5,352 participants, and the equations were validated in 4,050 participants. At least 85% of estimates in the validation data were within 30% of measured GFR. The combined race-free equation was more accurate than the new single-marker equations and showed smaller differences between Black and non-Black participants.

Those results support the equation change and the value of complementary markers. They also show why an eGFR should not be presented as exact: “within 30%” is an accuracy benchmark, not a promise that every estimate sits close to measured GFR.

Creatinine can carry information unrelated to filtration

Creatinine reflects filtration, but its blood concentration also depends on how much creatinine is generated and handled outside glomerular filtration. That is why a value can be a less reliable filtration marker in some contexts.

The National Kidney Foundation's laboratory implementation guidance lists altered creatinine generation in settings such as muscle wasting, amputation, or bodybuilding, and medicines that affect tubular secretion, among the non-GFR factors clinicians may consider. These are examples of context, not a home differential: one of them cannot be assigned as the cause of a result from the number alone.

Cystatin C is useful because its non-GFR influences differ from creatinine's, not because it is a perfect marker. NKF notes that smoking, obesity, inflammation, and thyroid or adrenal-hormone disorders can affect cystatin C. NIDDK therefore describes combined creatinine–cystatin C estimation as generally more accurate while still asking clinicians to consider which marker has important non-GFR determinants in a particular person.

A clinician may consider cystatin C, a combined estimate, or measured GFR when creatinine is less reliable or an estimate sits near a consequential decision point. That is a clinical judgment, not a universal testing schedule or an instruction created by this article.

KDIGO GFR categories are not universal “normal” ranges

The 2024 Kidney Disease: Improving Global Outcomes guideline defines six GFR categories for classifying chronic kidney disease:

• G1: 90 or higher mL/min/1.73 m², described as normal or high.
• G2: 60–89, described as mildly decreased relative to a young-adult level.
• G3a: 45–59, mildly to moderately decreased.
• G3b: 30–44, moderately to severely decreased.
• G4: 15–29, severely decreased.
• G5: below 15, termed kidney failure in the guideline.

These are guideline categories, not a portable creatinine reference interval, a universal diagnosis for one reading, or a personal treatment target. KDIGO specifically says that G1 and G2 do not meet chronic kidney disease criteria in the absence of evidence of kidney damage. Conversely, a higher eGFR does not exclude kidney disease when qualifying kidney-damage evidence is present.

The complete category also does not stop at “G3” or another G label. KDIGO uses CGA: cause, GFR category, and albuminuria category. That framework keeps a filtration estimate from standing in for the disease cause or the urine evidence used to assess kidney damage and risk.

The broader guide to understanding lab reference ranges explains why a laboratory interval, a guideline classification boundary, and a personal clinical decision point answer different questions.

One result cannot establish chronicity

Chronic kidney disease is defined by an abnormality of kidney structure or function present for at least three months with implications for health. The KDIGO 2024 guideline says to repeat an incidentally detected low eGFR, elevated urine albumin-to-creatinine ratio, or hematuria to confirm the presence of CKD. It also says not to assume chronicity from one abnormal eGFR or ACR because the finding could reflect recent acute kidney injury or acute kidney disease.

Proof of chronicity can come from prior GFR estimates, earlier urine findings, imaging or pathology, relevant history, or repeat measurements within and beyond the three-month point. The appropriate route and timing depend on the clinical situation. “Three months” is part of the disease definition; it is not a one-size-fits-all instruction to wait three months before seeking review.

This creates two separate questions:

1. What does the current result show? A measured marker and a calculated estimate at a particular collection time.
2. What has been established over time? Persistence, prior kidney evidence, or a recent acute change evaluated in clinical context.

Do not flatten those questions into “the trend is the diagnosis.” A sequence must first be source-compatible, and chronicity still does not identify the cause, urgency, or treatment.

Urine albumin adds a different kidney signal

eGFR describes estimated filtration. Urine albumin looks for evidence that albumin is passing into urine. NIDDK's guide to assessing urine albumin identifies the urine albumin-to-creatinine ratio, or UACR, as the recommended spot-urine measure for assessing and monitoring albuminuria.

The “creatinine” in UACR is urine creatinine used to relate albumin to urine concentration. UACR is not the serum creatinine result used in eGFRcr, and urine albumin is not the same analyte as the blood albumin reported on a metabolic or liver panel.

KDIGO recommends using both urine albumin measurement and GFR assessment in people at risk for or living with CKD. Either dimension can add information the other does not contain. Preserve the exact urine test name, value, unit, source range or category, specimen, and date instead of recording only “protein normal” or “kidney labs okay.”

Equation changes can look like biological changes

Before describing two eGFR values as a trajectory, compare the source labels. The NKF implementation guide warns laboratories that results from the 2021 equations do not trend directly with values from older equations. A change in the reported value can therefore reflect a change in equation, marker inputs, laboratory implementation, or source context rather than a proven change in the person's filtration.

Check whether both reports used:

• the same analyte and unit for serum creatinine;
• the same eGFR label and equation;
• creatinine alone, cystatin C alone, or both markers;
• the same indexed unit of mL/min/1.73 m²;
• comparable laboratories, methods, and collection contexts.

The workflow for combining lab results from different laboratories shows how to label a sequence as comparable, qualified, or unresolved. Keeping results separate until the evidence supports comparison is more honest than drawing a smooth line through incompatible rows.

Keep the measured marker, the estimating method, and the clinical question attached. A kidney number without its provenance is missing part of its meaning.

A five-step way to prepare creatinine and eGFR for review

1. Preserve the measured result

Keep the exact serum creatinine name, value, unit, flag, report interval, laboratory, collection date, specimen, and any method or sample note. Save the source report when possible so the value remains attributable.

2. Name the estimate

Record the eGFR value and indexed unit exactly as reported. Preserve the equation or marker label: eGFRcr, eGFRcys, eGFRcr-cys, CKD-EPI 2021, or another stated method. Note the equation's age scope rather than applying an adult formula to a child or assuming every laboratory used the same calculation.

3. Keep the source context

Preserve why testing was ordered, relevant history, medicines, body-composition context, symptoms, and the clinician's question. These facts can affect how a marker or estimate is evaluated, but the record should not assign one as the cause.

4. Add chronicity and kidney-damage evidence

Keep prior creatinine and eGFR reports, UACR or other urine findings, urinalysis, imaging, and other clinician-selected evidence visible. Label whether the record actually establishes persistence beyond three months instead of inferring it from one collection.

5. Check comparability and prepare questions

Before calling a change a trend, compare analyte, unit, laboratory, equation, marker inputs, and context. Then prepare questions such as:

• Which eGFR equation and filtration marker or markers produced this result?
• Are there reasons creatinine or cystatin C may be less reliable in this context?
• What urine, historical, structural, or other evidence belongs with this estimate?
• Does the available record establish chronicity, or is a recent change still being evaluated?
• Is the estimate close to a clinical decision where additional precision matters?

The complete guide to reading blood-test results keeps source facts, comparison, interpretation, and treatment decisions in separate lanes.

A source-aware example

Suppose one report shows serum creatinine and an eGFRcr flag. A bounded summary might say: “Serum creatinine value, unit, interval, laboratory, date, and sample notes preserved; eGFRcr recorded with its equation label; prior kidney results and urine albumin evidence not yet assembled; chronicity and cause not established.”

If an older report contains a different eGFR, first check the equation and marker labels. The two reports may document different estimates without proving a biological change. Add the missing source facts before asking a clinician what the sequence means.

Where Libby fits

Libby can support the organization layer: keeping source reports, exact serum creatinine values, units, eGFR labels, dates, urine results, related records, and context notes together. It does not measure GFR, determine which equation should apply, decide whether markers are reliable, reconcile laboratory methods, diagnose acute or chronic kidney disease, assess urgency, prescribe testing, or recommend treatment.

Use the organized record to prepare the clinician questions above. If that workflow would help, you can start your record.

FAQ

Are creatinine and eGFR the same test? No. Serum creatinine is a measured concentration. eGFR is a calculated estimate that uses creatinine or cystatin C, often with age and sex. Preserve both results and the equation label rather than treating them as duplicate values.

Does one low eGFR mean chronic kidney disease? Not by itself. KDIGO defines chronic kidney disease using abnormalities present for at least three months and warns against assuming chronicity from one abnormal eGFR or ACR. A recent acute change and a chronic finding require different clinical inferences.

Is an eGFR above 60 always normal? No universal conclusion follows from that number alone. KDIGO says G1 and G2 do not meet CKD criteria without kidney-damage evidence, but qualifying albuminuria, structural findings, or other markers can still matter at a higher eGFR.

Why might a clinician consider cystatin C? Cystatin C has different non-GFR influences from creatinine. Combining the two markers generally improves eGFR accuracy, especially when a creatinine-only estimate may be less reliable or sits near a consequential decision point. Cystatin C still has limitations and is not automatically required for everyone.

Why does UACR matter if I already have eGFR? They provide different evidence. eGFR estimates filtration, while UACR assesses urine albumin relative to urine creatinine. KDIGO's CKD framework uses both GFR and albuminuria rather than treating either result as the whole picture.

Can I compare eGFR results from different laboratories or years? Only after checking the serum marker, unit, eGFR equation, marker inputs, laboratory, and source context. A laboratory's switch to a newer equation can change the reported estimate without proving a biological change.

Does creatinine make eGFR a biological-age score? No. Age is an input in adult estimating equations, but eGFR is an estimate of glomerular filtration for a defined clinical use. It is not a biological-age verdict, longevity target, or instruction to change treatment.

References

• KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease
• eGFR Equations for Adults — NIDDK
• Chronic Kidney Disease Tests and Diagnosis — NIDDK
• Assess Urine Albumin — NIDDK
• New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race — PubMed Central
• Recommendations for Implementing the CKD-EPI 2021 Race-Free eGFR Calculation — National Kidney Foundation
• National Kidney Foundation Laboratory Guidance for CKD-EPI 2021 — PubMed