How to track your eGFR over time

eGFR is on almost every basic or comprehensive metabolic panel, so most adults have dozens of them without ever having asked for one. It is the standard estimate of how well your kidneys filter, and it is one of the tests where a history matters most. A single eGFR of 62 says little on its own. Ten years of results show whether that 62 is where you have always been, or the latest step in a slow slide from 90.

It is also one of the easiest tests to misread over time, because the number is calculated, and the calculation has changed more than once.

What eGFR estimates

GFR, the glomerular filtration rate, is how much blood your kidneys filter each minute. Measuring it directly takes an injected tracer and timed samples, so it is rarely done. Instead, labs estimate it from a blood test.

The usual starting point is creatinine, a waste product of muscle that the kidneys clear. When filtration falls, creatinine builds up. The lab plugs your creatinine, age, and sex into an equation and reports the result as eGFR, the estimated GFR, in mL/min/1.73m². The “1.73m²” scales the result to an average body size, so people of different sizes can be compared.

Because eGFR comes from creatinine, anything that changes creatinine without changing your kidneys also changes eGFR. That is the thread running through most of this post.

Cystatin C

Some labs also measure cystatin C, a protein made by nearly every cell in the body, not just muscle. It is not affected by muscle mass the way creatinine is. The KDIGO 2024 chronic kidney disease guideline recommends an eGFR calculated from both creatinine and cystatin C when the creatinine estimate is likely to be less accurate, such as in very muscular or very frail people, or when the result will drive a treatment decision. If your report shows “eGFR cystatin C” or “eGFR creatinine-cystatin C”, that is a different estimate from the usual one. Keep it as its own line.

The equation changed

There have been three main equations in U.S. labs in the last twenty years:

• MDRD. The older equation. It was less accurate at normal and near-normal kidney function, which is why labs using it often printed “>60” instead of a number.
• CKD-EPI 2009. More accurate at higher levels, and it replaced MDRD at many labs. Like MDRD, it included a multiplier for patients identified as Black.
• CKD-EPI 2021. A refit of the 2009 equation without race. A joint task force of the National Kidney Foundation and the American Society of Nephrology recommended it for immediate use, and the NKF now recommends it as the standard for adults.

For tracking, what matters is that the same creatinine gives a different eGFR under each one. A study of more than 14,000 results found the 2021 equation ran about 2% higher than the 2009 one on average, enough to move one result in ten into a different CKD stage. For Black patients whose earlier results were calculated with the race multiplier, the change often runs the other way, and can be larger.

So a history that spans the switch can show a step with no change in your kidneys. The NIDDK says to compare results calculated with the same equation, and asks labs to note the date they switched. The equation is often named in the test name or a footnote: “eGFR CKD-EPI 2021”, “eGFR (MDRD)”, or “eGFR non-African American”. When it is not, ask the lab, or note the first report where the number stepped without a matching change in creatinine.

If you have both the creatinine and the eGFR for an older result, the creatinine is the measured value. It has not changed meaning. Tracking creatinine alongside eGFR is one way to see through an equation switch. Another is to recalculate the older results with the eGFR calculator, so every point uses the 2021 equation.

The numbers on the report

eGFR cutoffs come from the KDIGO guideline, and they are the same at every lab:

A few things about this table are easy to miss.

• One low result is not a diagnosis. Chronic kidney disease means abnormal kidney structure or function for at least three months. The guideline says not to assume it from a single eGFR, which could reflect a short-term injury that resolves.
• G2 on its own is not kidney disease. An eGFR of 60 to 89 counts as CKD only when there is other evidence of kidney damage, most often albumin in the urine.
• eGFR is half of the staging. KDIGO also grades albuminuria, using the urine albumin-to-creatinine ratio (ACR, sometimes UACR): A1 is below 30 mg/g, A2 is 30 to 300, and A3 is above 300. An eGFR of 70 with an ACR of 400 carries more risk than an eGFR of 55 with a normal ACR. If you have urine ACR results, track them too.

“>60” and other capped values

Some labs, especially those still using MDRD, print “>60” or “>90” instead of a number. That is a floor, not a value. A history of “>60, >60, >60, 58” may look like a sudden drop, when the earlier results could have been 61 or 95. NIDDK recommends that labs now report exact whole numbers above 60, so newer reports usually do. When you record a capped value, keep the ”>” with it, and do not read the line between a capped value and a real number as a trend.

How often it is tested

• Routine care: eGFR comes along with most metabolic panels, so many people have one every year or two without anyone ordering it specifically.
• Diabetes: the American Diabetes Association’s Standards of Care recommend eGFR and urine ACR at least once a year in type 2 diabetes, and in type 1 diabetes after the first five years.
• Known CKD: KDIGO recommends both tests at least once a year, and more often for people at higher risk of progression or when the result will change treatment.
• After starting certain medications: often within a few weeks. More on that below.

Because eGFR rides along with routine panels, you may have more history than you think, spread across years of physicals at different clinics. Older results are the baseline that tells you whether today’s number is new. How to get all of your medical records covers tracking them down.

What counts as a real change

eGFR varies from draw to draw even when your kidneys are stable. KDIGO reports within-person biological variation of about 5%, and puts the threshold for a change that is probably real at around 14% to 20% between two results. On an eGFR of 70, a result anywhere from about 60 to 80 can be ordinary variation.

KDIGO’s practical rule for people with CKD is that a change of more than 20% on a later test exceeds the expected variability and is worth a closer look. On a baseline of 70, that means a result below 56. A drop from 70 to 64 is about 9%, which is within the noise for one pair of results.

A few questions help sort it out:

• Is the change bigger than your usual spread? If your last five results ran from 66 to 76, a 68 is not news. A 54 is.
• Did it hold? A single low value that recovers on the next draw is often a short illness, dehydration, or one of the factors below. A slope that keeps going is different.
• Did creatinine move, or only the equation? Check the creatinine and the equation name before trusting a step.
• What is the slope over years? Kidney function declines slowly with age for many people, though KDIGO notes the rate varies widely and some people show little decline at all. A long, gentle slope is different from a recent steep one, and only a long history can show which you have.

Things that move eGFR

Most of these work by changing creatinine, not filtration.

• Muscle mass. More muscle makes more creatinine and lowers eGFR. Bodybuilders can read low, and people who have lost a lot of muscle can read falsely high.
• Creatine supplements and high-protein diets. KDIGO lists both as reasons the creatinine estimate can be off.
• A cooked meat or fish meal. KDIGO cites a study where a meat meal raised creatinine by about 0.23 mg/dL, enough to lower eGFR by about 20 points in that group, with the peak 2 to 4 hours after eating. The guideline suggests waiting 12 hours after meat or fish before the draw, if you can.
• Intense exercise shortly before the draw.
• Some medications block the kidney’s secretion of creatinine, which raises creatinine without changing filtration. Trimethoprim, an antibiotic, is one that KDIGO names. Cimetidine and some HIV medicines do the same. The effect is reversible.
• Dehydration or an acute illness can lower real filtration for a while. That is a genuine change, but usually a short one.
• Blood pressure and kidney-protective drugs. ACE inhibitors, ARBs, and SGLT2 inhibitors often cause an early dip in eGFR in the first weeks, because they lower pressure inside the kidney’s filters. KDIGO treats a dip of up to about 30% as expected and generally not a reason to stop, and advises continuing an ACE inhibitor or ARB unless creatinine rises more than 30% within four weeks. A larger drop calls for a look at other causes. If you started one of these drugs, the result a few weeks later is not your new baseline yet.

None of these mean the result is wrong. They mean it needs a note next to it, and your clinician is the one to sort out which applies. Did the medication actually do anything? covers reading a before-and-after fairly.

What to record with each result

For each eGFR, keep:

eGFR shows up under many names: eGFR, Estimated GFR, Glomerular Filtration Rate, eGFR CKD-EPI, eGFR non-African American, eGFR African American, and eGFR if Black. The last few are the race-based results from before 2021, and some reports printed both on every panel. Pick one name for the result you track and use it everywhere, or your history will split into short, separate lines.

Charting it

A simple line chart works. A few choices make it honest:

• Space points by date. A slope only means something when the time axis is real.
• Mark the equation switch. A note on the date your lab changed equations keeps a step from looking like a decline.
• Chart creatinine alongside it. If creatinine is flat and eGFR stepped, it was the equation. Creatinine runs the opposite direction from eGFR, so read the shapes inverted.
• Do not connect capped values. A line from “>60” to 58 is not a trend.
• Mark medication changes, and treat the few weeks after a new ACE inhibitor, ARB, or SGLT2 inhibitor as settling time.
• Look at years, not pairs. The question for eGFR is usually the slope, and a slope needs many points.

How to track your lab results over time has the general method for any test. If you have diabetes, your A1c belongs on the same timeline, since diabetes is a leading cause of kidney disease.

Tracking eGFR in HealthViewer

HealthViewer is a lab results tracker that runs on your own computer. It imports eGFR from lab report PDFs, MyChart downloads, and direct connections to your health system. It files eGFR, Estimated GFR, Glomerular Filtration Rate, and eGFR non-African American under one test, along with the standard codes health systems use for the MDRD, CKD-EPI, and CKD-EPI 2021 results. The race-specific “African American” and “Black” lines are deliberately kept separate. A name like “eGFR CKD-EPI 2021” from a PDF may land as its own test, and you can combine it with the main one using Merge in the Test Catalog.

Because the three equations share one series, an equation switch can show as a step on the chart. A journal note on the date your lab switched keeps it visible, and journal entries can appear as markers on the chart.

In Trends, the Kidney tag pulls together BUN, creatinine, and eGFR, and eGFR is also in the Metabolic Panel filter. The overlay view puts eGFR and creatinine on one chart, and medication markers show where a blood pressure drug or SGLT2 inhibitor started.

Everything stays in an encrypted file on your computer, with no account and no cloud service. Getting Started walks through the first import.

This post is general information, not medical advice. Talk to your clinician about what your eGFR means for you and before changing any treatment.