Eliminating race in calculating renal function

 Several organizations have strongly supported eliminating race-based approaches to evaluating the level of chronic kidney disease through the estimated Glomerular Filtration Rate (eGFR), including the National Kidney Foundation/American Society of Nephrology (see ckd egfr not use race ajkd2021 in dropbox, or https://www.ajkd.org/action/showPdf?pii=S0272-6386%2821%2900828-3 , for the following reasons (embellished a bit by me):

-- race is a social and not a biological construct, given the significant genetic diversity in the overall population independent of skin color

-- any biological differences attributable to race are likely related to other factors, and attribution to race reinforces racial profiling/systematic racism and undercuts understanding of the complex social issues (and decreases the likelihood of studies actually including and addressing these important social determinants of health):

    -- race in our society, as in many others, is associated with an array of social conditions that may well be the causes of apparent racial differences of disease, such as chronic stress on a daily basis felt by many of those with darker skin (and chronic stress is associated with increased inflammatory markers, which are associated with many adverse health outcomes, as well as higher cortisol and catecholamine levels, which affect almost all of the other hormones in the body and their targets), differences in diet and access to healthy foods, differences in living conditions and housing quality, differences in financial stresses, etc., etc.

-- for this report, the Task Force identified 26 approaches for the estimation of GFR (eGFR) that did or did not include race, then narrowed their focus by consensus to five of these approaches

 

Recommendation #1: The Task Force recommends for U.S. adults (>85% of whom have normal kidney function) that the CKD-EPIcr_R retrofit that was developed without the use of the race variable be implemented immediately, including all laboratories. In addition to not including race in the calculation and reporting, it included diversity in its development, is immediately available to all labs in the U.S., and has acceptable performance characteristics and potential consequences that do not disproportionately affect any one group of individuals.

 

Recommendation #2: The Task Force recommends national efforts to facilitate increased, routine, and timely use of cystatin C, especially to confirm eGFR in adults who are at risk for or have chronic kidney disease. Combining filtration markers (creatinine and cystatin C) is more accurate and would support better clinical decisions that either one marker alone. Thus, if ongoing evidence supports acceptable performance, the CKD-EPIcys [based on cystatin C] and CKD-EPIcr-cys_R [based on combo of creatinine and cystatin C] without the race variable should be adopted to provide more accurate first line or confirmatory testing as appropriate for the clinical setting.

 

Recommendation #3: The Task Force recommends that research on GFR estimation with new endogenous filtration markers and on interventions to eliminate race and ethnic disparities in kidney disease be encouraged and funded. We implore American society-at-large and funding agencies to invest in developing the science needed for new approaches for accurate, unbiased, and precise GFR estimation, as well as for improving estimation of physiologic function in other areas of medicine, with the ultimate goal of promoting health equity.

 

Commentary: 

-- it is certainly a welcome change to eliminate pigeonholing people based on skin color, and actually (hopefully) assessing the actually important sociodemographic features of the patients involved in future studies

    -- this approach would hopefully open doors to dealing with these sociodemographic factors that are associated with differences in clinical outcomes

-- there are still, of course, racial biases in medicine and outcomes: Black patients have worse outcomes with respect to blood pressure control, timely nephrology referrals, fistula placement prior to hemodialysis initiation, waitlisting for kidney transplantation, and receiving a transplant

-- there are clearly adverse effects to having inaccurate estimations of GFR, including affecting advisability of contrast material/cardiac catheterization, use of medications that are affected by creatinine clearance (e.g. metformin or SGLT2 inhibitors), appropriate dosing for pain or cancer medications, inclusion in clinical trials (many exclude patients having low eGFRs), and perhaps other clinical decision-making including such things as kidney donor acceptability.

-- Their focus on cystatin C as an alternative to creatinine was because it is expressed in all nucleated cells, produced at a constant rate, and freely filtered by the glomerulus. Unlike creatinine, it is not secreted but is instead reabsorbed by tubular epithelial cells and catabolized and not returned to the bloodstream. This assay is reasonably widely available, unlike some newer filtration markers, and there are some high throughput analyzers of cystatin C and standardization is available.

 

-- in this light, a new large national study of adults with CKD found that using cystatin C vs creatinine as marker of renal function eliminated the need for including race in estimating GFR (see ckd cystatin c without race nejm2021 in dropbox, or DOI: 10.1056/NEJMoa2103753). 

    -- this study evaluated 1248 participants who had CKD and available information on race (reported by patient), genetic ancestry markers, serum creatinine, serum cystatin C, and 24-hour urinary creatinine levels. They compared the calculated eGFR vs measured GFR (by urinary ¹²⁵I-iothalamate clearance) in those self-identified as Black vs non-Black

    --median age 56, 57% male, 36% Black/48% white, 12% college grad/15% professional/23% some college/19% high school (with increasing percentages achieving higher education levels in the “non-Black” group) 

--genetic ancestry: African 31% (80% of self-identified Black/2% non-Black); European 56% (15% of self-identified Black/96% non-Black) 

--serum creatinine 1.7 mg/dL (1.8 in Black participants, 1.6 in non-Black); serum cystatin C 1.45 mg/L (1.47 in Black participants, 1.44 in non-Black): both of these differences were not statistically significant; mean urinary protein level: 0.2 g/24h (0.3 in Black participants, 0.2 in non-Black), p=0.04 (patients who self-identify as Black on average have higher serum creatinine levels, independent of age, sex or measured GFR (ie they have less actual renal failure at the same GFR as those who self-identify as “non-Black”)     

        -- for Black persons: not using race in the equation with creatinine led to under-estimation of eGFR by 3.99 ml/min/1.73 m² body surface area (2.17-5.62) and lower accuracy (% of eGFR within 10% of measured) by 31% (24-39%), when compared to measured GFR 

            -- for non-Black: creatinine-based eGFR was off by -0.92 (-2.29 to 0.55), ie over-estimation, and accuracy 34% (29-40) when compared to measured GFR 

        -- incorporating genetic ancestry data instead of race: under-estimation of eGFR by 1.33 ml/min/1.73 m² BSA (-0.12 to 2.33) and lower accuracy by 42% (34-50%): not statistically different from using race 

        -- inclusion of non-GFR determinants of serum creatinine level (body composition metrics, urinary excretion of creatinine) that differed by reported race and genetic ancestry did not eliminate misclassification by removing race

        -- BUT: eliminating race or genetic ancestry and using cystatin C measurement instead of creatinine did not affect the eGFR: mean difference in eGFR vs measured in Black persons was 0.33 (-1.43 to 1.92) or accuracy at 41% (34-49%); for non-Black persons, mean difference between measured and estimated GFR was 0.29 ml/min/1.73 m² BSA (-0.84 to 1.36) 

    --so, bottom line from this study: 

        --including self-identified Black race or measured genetics by ancestry led to a significant difference in measured vs calculated GFR when using creatinine levels (and the level of this difference at around 13% would be sufficient to affect clinical decision-making, as noted above) 

        --but the eGFR by cystatin C does not vary by race/genetics (ie, there are lower rates of creatinine tubular secretion in Black/African heritage persons, not found with cystatin C) 

        --and these differences using creatinine was not more accurate when using non-GFR determinants of serum creatinine such as body metrics in the calculation  [though, one issue with body metrics is that creatinine derives from muscle breakdown; so, using body surface area (BSA), which includes weight and height, may not truly reflect muscle mass. and, persons of color tend to have more manual jobs and may well on average have more muscle mass)

  

Another study was just published in NEJM evaluating prior studies, deriving a new and more accurate eGFR equation incorporating both creatinine and cystatin C but not race (see ckd new eGFR equation incl creat and cystatin nejm2021 in dropbox, or DOI: 10.1056/NEJMoa2102953). 

--data from 2 development data sets: 10 studies (8254 participants, 31.5% Black) for serum creatinine and 13 studies (5352 participants, 39.7% Black) for both serum creatinine and cystatin C to develop a new eGFR equation not including race.  Validation data set of 12 studies (4050 participants, 14.3% Black) to compare the accuracy of the new eGFR equation with measured GFR 

--in validation data set: 

    --current creatinine-based eGFR using age, sex, and race: overestimated measured GFR in Black persons by 3.7 ml/min/1.73 m² BSA (1.8-5.4) and in non-Blacks by 0.5 ml/min/1.73 m² BSA (0-0.9) 

    --current creatinine-based eGFR using age, sex, but not including race: underestimated measured GFR in Black persons by 7.1 ml/min/1.73 m² BSA (1.8-5.4) and in non-Blacks by 0.5 ml/min/1.73 m² BSA (0-0.9) 

    --new equation based on creatinine and cystatin C but without race (only age and sex): underestimated measured GFR in Black persons by 3.6ml/min/1.73 m² BSA (1.8-5.5) but overestimated GFR in non-Blacks by 3.9 ml/min/1.73 m² BSA (3.4-4.4) 

        -- overall, at least 85% of the equations were within 30% of measured eGFR [for unclear reasons, they posit that up to a 30% margin of error “is considered adequate for clinical decision making in many circumstances”. I suppose this is true when people are mid-range in CKD categories, but lots of people seem to hover at the margins of these categories in my experience…..] 

    --new creatinine-cystatin C equations were more accurate than the new creatinine only ones 

    --the current cystatin C eGFR calculation is almost as good as the new creatinine-cystatin C equation in predicting measured GFR 

--so, bottom line of this study: 

    --their newly derived creatinine-cystatin C equation without incorporating race as a factor was quite close to the measured GFR 

    --the current cystatin C eGFR found essentially no difference in Black vs no-Black persons, but was a little less accurate than the combo creatinine-cystatin C one when compared to the measured GFR [which does raise the possibility that a new modeling might improve the accuracy of the eGFR with cystatin C alone] 

    --as with any new mathematical modeling equation, there should be verification prospectively in different situations with different people with different social/medical conditions before adopting it on a large scale. 

so, pretty impressive stuff. a few points:

--eliminating race from the eGFR calculations is really important for the reasons stated above: using race reinforces the concept that people are fundamentally different along racial lines and potentially leads to more stigmatization and reinforces racism, despite the fact that race is a social and not a biological construct; using race as a category undermines the very real issues of racial differences in our society that lead to poorer outcomes and these "racial" differences themselves need to be dealt with (increased daily stress from racism, decreased financial stability, decreased access to affordable healthy food, more stressful neighborhoods, decreased access to healthcare...... etc)

--the above recommendation to use the current eGFR calculator without including race is a first step: clearly there are many people near a specific cutpoint of renal function (eg, stop drug xxx if the eGFR <30, and their eGFR is 29) who might have a different eGFR if not using a creatinine-based calculator. but for the majority of people, the small differences do not lead to a significantly different approach to their care. and following the creatinine-based eGFR over time is a useful way to assess the evolution of the CKD (unless some medication were started that changes creatinine secretion, including sulfamethoxazole/trimethoprim, cimetidine/ranitidine, cephalosporins, fenofibrate....)

--the data from the study on a new calculator is intriguing, but needs to be tested further in diverse populations

--including cystatin C in the calculations makes lots of sense. it's been around a long time and seems much more accurate than creatinine (which also is muscle-mass dependent, unlike cystatin-C which is expressed by all nucleated cells)

--the study using just cystatin C without race is also compelling, though should have further research support

--the combination of both cystatin C and creatinine, per the recommendations, also makes sense at this time, and i think should be incorporated into our clinical practice

--for access to the calculators, go to https://www.kidney.org/professionals/kdoqi/gfr_calculator 

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list