New eGFR equations without race
The New England Journal of Medicine published an article which presented new creatine- and cystatin C-based eGFR equations without race as a variable. Current eGFR equations include race (Black vs. non-Black), beside age and sex. Race was initially included in the equations because studies indicated that Black persons had higher average serum creatinine levels of the same measured GFR level than non-Black persons [1-3]. The authors of the current article state that race is a social and not a biologic construct and that incorporation of race in eGFR equations disregards diversity within and among racial groups. The authors of the study developed new eGFR equations that do not include race and compared the accuracy of current and new eGFR equations using a validation data set.
The study showed that new creatinine-cystatin C equations without race were more accurate than new creatinine equations without race.
The new creatinine-cystatin C equations minimally underestimated measured GFR in Black participants (median difference between measured GFR and eGFR: 0.1 ml/min/1.73 m²; 95% CI -0.1 to 1.6) and overestimated measured GFR in non-Black participants (median -2.9 ml/min/1.73 m²; 95% CI -3.3 to -2.5, difference between Black and non-Black participants: 3.0, 95%CI 1.6 to 4.4). The new creatinine equations without race underestimated measured GFR in Black participants (median: 3.6 ml/min/1.73 m², 95% CI 1.8 to 5.5) and overestimated measured GFR in non-Black participants (median: -3.9 ml/min/1.73 m², 95% CI -4.4 to -3.4, difference between Black and non-Black participants: 7.6, 95%CI 5.6 to 9.5). In comparison, the current guideline-recommended creatinine equations that include race as a variable overestimated measured GFR in Black participants (median: -3.7 ml/min/1.73 m²; 95% CI -5.4 to -1.8) and minimally overestimated measured GFR in non-Black participants (median: -0.5 ml/min/1.73m², 95% CI -0.9 to 0.0, difference between Black and non-Black participants: -3.2, 95%CI -5.0 to -1.3).
For all studied equations, ≥85% of the estimated eGFRs were within 30% of the measured GFR, which is in many cases considered as an acceptable margin of error for clinical decision making.