Population-based studies have found that Black Americans with renal tumors are less likely to undergo a partial nephrectomy (PN) compared to non-Black Americans despite increased rates of comorbidities and risk of chronic kidney disease (CKD).^1–4 Guidelines highlight how preoperative renal function, tumor size, location, and complexity should be considered in the decision to perform PN or radical nephrectomy (RN).^5,6 However, creatinine-based equations to predict renal function are racialized and result in a 15%–20% higher estimated glomerular filtration rate (eGFR) if the patient is Black compared to non-Black patients.^7,8 Also, race adjustment is problematic since race is a social rather than a biological construct and may be subject to significant bias, especially if assumed by medical staff. The American Society of Nephrology and the National Kidney Foundation formed a task force in August 2020, which recommended the discontinuation of these equations.⁹ We hypothesized that these racialized estimates of eGFR may be one of many contributing factors to the lower utilization of PN in Black patients.

Figure 1. Multivariate analysis demonstrating association between partial nephrectomy and patient characteristics in full cohort (a, n = 6,327) and subgroup of patients with nephrometry scores (b, n = 2,663). BMI indicates body mass index; CCI, Charlson Comorbidity Index; CI, confidence interval; eGFR, estimated glomerular filtration rate.

Figure 2. Partial nephrectomy (PN) rates among chronic kidney disease (CKD) groups in Black and non-Black patients.

From 2005 to 2020, we included 6,327 consecutive patients who had undergone either PN or RN and had perioperative creatinine values, with available self-reported race at the Cleveland Clinic. We excluded patients with a tumor thrombus (n = 149). Data were dichotomized into Black (n = 654) and non-Black patients (n = 5,673), and compared across preoperative variables and surgery types. Multivariable regression analysis was used to identify predictors of PN. We also stratified PN rates across CKD stages and nephrometry complexity score groups for Black and non-Black patients. We compared the impact of using the race-modified (2009 Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI]) and race-free (2021 CKD-EPI refit) equations to estimate glomerular filtration rate and the changing distribution of patients across CKD stages. All P values < .05 were considered statistically significant.

In our cohort, 10.3% of patients were Black, and 38.7% were female. Compared to non-Black patients, Black patients had higher rates of diabetes, hypertension, comorbidity, and lower baseline eGFR before surgery (P < .001 for all; Table 1). On multivariable logistic regression analysis, Black patients were less likely to be treated via PN (OR=0.76, CI=0.61–0.96; P = .001), even when controlling for body mass index, age-adjusted Charlson Comorbidity Index, hypertension, preoperative eGFR, and clinical tumor size (Fig. 1). In the subset of patients (42%) with a R.E.N.A.L. (for radius, exophytic/endophytic, nearness of tumor to collecting system, anterior/posterior, location relative to polar line) nephrometry tumor complexity score (n = 2,663), Black patients had also a 37% lower odds of receiving PN than non-Black patients on multivariate analysis (OR = 0.63, CI=0.42–0.97; P < .001). There was no disparity in PN utilization in settings where the AUA guidelines recommend its prioritization. Utilization of PN was similar between Black and non-Black patients when stratified according to tumor complexity and CKD 2-5. The only significant difference was noted in the CKD1 group, where 55.84% of Black patients underwent PN compared to 67.43% of non-Black patients (P = .0011), thus driven by the presence of normal kidney function and presumably very low risk for postoperative CKD3b or higher, in which case the AUA guidelines indicate that RN is reasonable (Fig. 2). When comparing the preoperative eGFR values obtained through CKD-EPI and CKD-EPI refit, a drop was noticed in Black patients (73.6 [49.7, 95.2] vs 65.7 [44.1, 83.9]), while an increase was noted in non-Black patients (76.8 [58.4, 92.2] vs 78.7 [60.1, 94.4]). Also, the use of the race-free equation was associated with a shift to a higher CKD stage (worse kidney function) for 27% of Black patients (Table 2 and Fig. 3). Forty-eight of the 420 Black patients (11.4%) initially considered having a normal kidney function (CKD1-2) would have an abnormal kidney function (CKD≥3a) when using the race-free equation. Also, when omitting the race factor, the prevalence of CKD ≥3a rose by 7.3% in Black patients. Due to lower preoperative predicted renal function using the race-free equation and a lower percentage receiving PN in Black patients, the postoperative kidney function estimates were significantly lower than non-Black patients, with larger differences seen when using the race-free equation (59.2 [37.7, 77.8] vs 60.0 [44.5, 77.7] for Black and non-Black patients respectively; P = 0.02 using CKD-EPI and 54.7 [35.5, 71.2] vs 64.4 [48.2, 83.0] for Black and non-Black patients respectively; P < 0.001 using CKD-EPI refit). Limitations included a single-center retrospective design and unknown confounders of the surgery type choice.

Figure 3. Chronic kidney disease (CKD) staging using race-modified and race-free equations in Black (A) and non-Black patients (B, P < .001). CKI-EPI indicates.

The race-modified eGFR may lead to overestimation of renal function and contribute to clinical decision making toward more radical kidney surgery in some Black patients. Racialized kidney function equations, along with other unknown factors, may contribute to the disparity seen in the use of kidney preservation strategies for Black patients with renal masses and should be reconsidered.