Lewis Whyld/Press Association via AP Images
A kidney patient receives treatment on a dialysis machine.
Kidney doctor Carmen Peralta was having a normal day at her office last year when a patient asked to see her.
The man walked in holding a printed sheet of test results he'd received bearing two numbers. Each represented his baseline kidney function. One was within normal range; the other was not.
“He walked into my office and said, ‘Well, Doctor, which one am I?'” Peralta said.
The test—meant to measure the man's estimated glomerular filtration rate, or eGFR—came back with two results by design. For the past 20 years, kidney doctors and primary-care practitioners have agreed to use a formula for eGFR that gives split results for African American and white patients, in an attempt to get more accurate measurements.
Peralta's patient, a mixed-race American of Dominican ethnicity, told her he was unsure where this left him. At the time, she didn't have an answer.
Its name may be obscure, but eGFR is far from an unusual measurement. Michael Shlipak, who heads the Kidney Health Research Collaborative at the University of California-San Francisco, estimated that American doctors perform over 200 million eGFR tests each year. Whether they know it or not, almost every inpatient and every outpatient over 50 years old will receive one during routine blood tests.
Increasingly, though, doctors like Peralta and Shlipak are beginning to question eGFR in ways that stretch beyond patients' confusion. They both said the test is imprecise for patients of all races, leading to shaky results in as many as a third of patients within critical range.
“If you are underdiagnosing kidney disease, you can let patients take medications that may harm their kidney because you don't have a correct estimate,” Shlipak said in an interview.
“When you overdiagnose kidney disease, you cause patients stress, you tell them they can't take ibuprofen when they're in pain. So maybe they're more likely to take opiates or some other alternative because they think it's unsafe for them to take something safer.”
For African American patients in particular, the separate results can lead to underdiagnosis of chronic kidney disease, longer transplant wait times, and ill-advised treatment plans. Giving them different results was intended to make the test better for everyone. In reality, though, it exacerbates problems they already face in getting quality care.
How a Kidney Test Compensates for Race
To understand how the eGFR formula became so common, it's important to get a sense of how the test, and the kidneys more broadly, work.
Well-functioning kidneys filter blood by passing it through a cluster of blood vessels. The walls of that cluster—the glomerulus—are porous enough to drain out small chemicals that aren't useful elsewhere in the body, like urea and wayward ions.
In order to test kidney function, doctors can measure the level of one of those chemicals—creatinine. If the glomerulus isn't filtering it out at the right rate, creatinine builds up in the bloodstream. Patients with high creatinine levels in their blood likely aren't filtering well in general.
First discovered in 1847, creatinine levels in the blood are easy to measure and to understand. This made the chemical a good candidate for use in a test meant to be used every day, at nearly every hospital in the country.
Creatinine and eGFR don't have a one-to-one relationship, though. Because it's produced at a constant rate in muscles, patients with higher muscle mass have higher creatinine levels. Even if their kidneys are actually filtering at the same rate, a bodybuilder and a gym-phobic would have very different creatinine levels in their blood.
More than a decade ago, a group of doctors working at the Tufts Medical Center came up with a solution: an equation that would allow doctors to take creatinine levels, pass them through a simple equation, and come up with a number between 0 and 120 that represented actual filtration rate. A score of zero meant the kidney filtered little to no waste product, and a score of 120 meant it functioned perfectly.
The researchers decided to solve the muscle mass problem by including a small number of corrections in the formula. Because women and older patients have less muscle mass on average, the formula raises their results.
African Americans have, on average, higher muscle mass than white Americans. The researchers decided to account for this, raising their eGFR by 15.9 percent.
They called their formula the CKD-EPI equation and published it in the Annals of Internal Medicine in 2009. Today, the National Kidney Foundation’s website terms it “the best overall index of kidney function.” An earlier version published in 2006 by several of the same researchers and still used in some hospitals also corrects for race. That formula raised African Americans’ results by 21 percent.
The group of doctors and epidemiologists that produced the CKD-EPI equation and its forerunner have cited robust studies about its effectiveness for the average patient.
But Peralta, Shlipak, and other doctors who believe eGFR isn't good enough, particularly for patients of color, point out that individuals don't conform to averages.
The Consequences of Bad Diagnoses
The issues with eGFR mean African American kidney patients face inequities at the earliest stages of their treatment, before they even know they may be at risk.
Despite their higher scores, Shlipak said African Americans are more likely than white patients to suffer from reduced kidney function.
“We know that blacks have a higher risk of reaching dialysis, they have more risk factors like hypertension, diabetes, obesity, so from a public health standpoint, it's a major problem,” he said.
For an African American who happens to have below-average muscle mass, a 16 percent or 21 percent bump in estimated kidney function can mean getting on a waiting list for a new kidney one year later. It can mean beginning dialysis well after a white patient would, or continuing to take ibuprofen when they likely shouldn't.
The direst consequences for these patients often come when they are seeking a live-donor kidney transplant. Generally, patients must have an eGFR below 20 or 21 before they are included on a transplant waiting list. The correction for African American patients can mean they get on a list months or years later than they otherwise might have.
“If your estimate, because you're African American, is GFR 23, but if you're white it's 19, that could definitely make a difference in terms of being listed or not listed,” Peralta said. “And it could be a year too much extra.”
Meanwhile, African Americans who need kidney transplants face additional hurdles.
In 2019, the U.S. Organ Procurement and Transplantation Network reported that over the past 30 years, just 12.6 percent of kidney donors in the U.S. were African American on average, meaning the chances of a good genetic match are lower. African Americans made up 31.9 percent of patients waiting for a kidney in 2019, and 27.1 percent of patients who actually received a transplant.
A 2018 investigation, published in the Journal of the American Medical Association, that examined efforts to reduce racial disparities in kidney transplants concluded that “national strategies to reduce disparities in [live-donor kidney recipients] have not been effective.”
Although the investigation's authors did not mention eGFR, they cited a number of “barriers” patients of color confront at every step, including initial screening.
But studies explicitly questioning eGFR are beginning to crop up with greater frequency. In June, a group of nephrologists working at the University of Pennsylvania published a note in JAMA that called into question the race correction. They criticized the use of race, now widely considered a social construct rather than a biological one.
“The use of kidney function estimating equations that include race as a variable cause problems for transparency and unduly restrict access to care in some cases, yet offer only modest benefits to precision,” they wrote.
A Troubling Stalemate
Despite the calls to reevaluate it, eGFR remains an entrenched part of the way doctors understand kidney disease.
Nephrologist Yoshio Hall, who sits on the board of the National Kidney Foundation, said he does not foresee a move away from the formula without a lot more research and education in the next several years. Colin Geddes, a nephrologist at the British National Health Service, said he thinks a move away from the test would be ill advised.
The researchers who see eGFR as fundamentally flawed are beginning to advocate for change, though. An alternative test using the protein cystatin C—which is less influenced by muscle mass—is often used when results from creatinine come back inconclusive.
Shlipak said he believes that test could easily become affordable. He currently sits on the board of Kidney Disease: Improving Global Outcomes, a collective of nephrologists from around the world. The group first recommended examining alternative tests in 2013 and will meet again in the fall to draft new international guidelines.
For Peralta and her patients, though, change needs to come more quickly.
“It's enormous,” she said. “It can be the difference between a diagnosis of chronic kidney disease or not, for which there's a whole new offering of care.”