Abstract
Utilization of Hepatitis C (HCV) viremic kidneys is increasing in the United States. We examined racial disparity in this utilization using UNOS/OPTN data (2014–2020) and mixed effects models adjusting for donor/recipient/center factors. Included in the study were 58,786 adults receiving a deceased-donor kidney transplant from 191 centers. 2,613 (4%) received kidneys from HCV-viremic donors. Of these, 1,598 (61%) were HCV-seronegative and 1,015 (49%) were HCV-seropositive. Among seronegative recipients, before adjusting for waiting time and education, Blacks (OR 0.69, 95%CI [0.60, 0.80]), Hispanics (OR 0.63, 95%CI [0.51, 0.79]) and Asians (OR 0.69, 95%CI [0.53, 0.90]) were less likely than Whites to receive HCV-viremic kidneys. In final models, effect of race was attenuated. Notably, shorter waiting time (OR 0.65, 95%CI [0.63,0.67]) and increasing educational level (Grade School less likely compared to High School OR 0.67, 95% CI [0.49, 0.92] and College more likely than High School (OR 1.16 95% CI [1.02, 1.31]) were associated with receipt of HCV-viremic kidneys. Among HCV-seropositive recipients, recipient race was not independently associated with receipt of HCV-viremic kidneys however, centers with larger populations of Black waitlisted patients were more likely to utilize HCV-viremic kidneys (OR 1.71, 95%CI [1.20, 2.45]) compared to other centers. Our results suggest recipient race does not independently determine who receives HCV-viremic kidneys however, other underlying factors including waiting time, education (among seronegative) and center racial mix (among seropositive) contribute to the current differential distribution of HCV-viremic kidneys among races.
Keywords: Race, Hepatitis C Viremic Kidney, Disparity
1. Introduction
Donor-derived disease significantly limits the number of organs available for transplantation.1 Some of these transmissible diseases include blood borne infections, specifically human immunodeficiency virus (HIV), Hepatitis B and Hepatitis C (HCV).2 Yet, while utility of organs from donors with HIV and Hepatitis B remain limited,3 contemporary advances in treatment efficacy of direct acting antiviral medication has allowed for the broader use of HCV-viremic organs.4,5 With treatment advances, HCV-viremic kidneys now represent a rich “new” resource augmenting the current kidney donor supply, and since 2014, Chang and colleagues have demonstrated a reducing trend in HCV-viremic donor kidney discard in the United States6.
While HCV-viremic kidney utilization increases nationally across the United States, it remains unknown how aggressively individual transplant centers use HCV-viremic kidneys, and which potential patients are selected to receive these kidneys. In this regard, centers may adopt different approaches. For example, all patients can be automatically listed to receive HCV-viremic kidneys and then consent obtained at the time of organ offer. Alternatively, centers can select patients based on specific criteria and obtain consent at initial transplant evaluation or wait-listing, interval evaluation, or at time of organ offer. Individual patient choices also play an important role. Unlike kidneys that do not carry the risk of infection, transplantation of HCV-viremic kidneys requires buy-in and explicit consent from patients. In the long run, choices made by centers and kidney patients will determine which patients are ultimately transplanted using HCV-viremic kidneys.
Race of an individual may influence clinical decision making and outcomes. Previous studies have demonstrated that Black patients express greater concern regarding the risks associated with kidney transplantation7 and have greater mistrust of the healthcare system.8 Mistrust may limit enthusiasm to receive HCV-viremic kidneys. Specifically, McCauley et. al. examined kidney transplant candidate willingness to accept HCV-viremic kidneys under hypothetical scenarios.9 They found, for Blacks compared to Whites, attenuated willingness to accept an HCV-viremic kidney in presented scenarios with varying HCV cure rates.9 Center racial mix and choices may also impact eventual outcomes. Racial differences in clinical outcomes have been well described in hospitals that disproportionately treat black patients compared to other centers.10 For example, Skinner et al found that risk-adjusted mortality following acute myocardial infarction was significantly greater at hospitals largely treating Black patients.11
We hypothesized that race may impact receipt of HCV-viremic kidneys. As such, this study examines whether patient-level racial differences exist in the receipt of HCV-viremic kidneys, while accounting for transplant center differences including center racial mix. Further, the study also aims to identify potential underlying reasons for any differences observed.
2. Materials and Methods
We conducted a retrospective cohort study of deceased donor kidney transplant recipients in the US between 12/4/2014 to 9/3/2020. The study was exempt from Institutional Review Board review.
2.1. Data Source
We used Organ Procurement and Transplantation Network (OPTN) data which has been previously well described.12
2.2. Study Population
Inclusion criteria were as follows: 1) Recipient age ≥18 years; 2) Deceased donor kidney; 3) First time transplant; 4) Transplant occurring between 12/4/2014 to 9/3/2020
2.3. Primary exposure
The primary exposure was race as defined by race and ethnicity variables in the Standard Transplant Analysis and Research (STAR) file. Race and ethnicity were combined into a single variable with the following categories: White, Black, Hispanic, and Asian.
2.4. Outcome Measure
The main outcome measure was receipt of an HCV-viremic kidney. (i.e., receipt of an HCV-viremic kidney is coded 1 while, receipt of any a deceased donor kidney which is HCV nucleic acid test negative is coded 0.
2.5. Candidate Covariates
Covariates were selected a priori based on review of the literature for factors that could influence receipt of a deceased donor kidney. We included three center level covariates: kidney transplant volume, center utilization of public health service (PHS) increased risk profile kidneys, and center race composition. Center volume was defined using methods described by Sonnenberg et al.13 Similarly, center utilization of PHS increased risk profile kidneys was defined and categorized based on quartiles of the ratio of the total number of PHS increased risk kidneys a center utilized to total volume of adult deceased donor transplants at that center performed during the study period. Categories included high, medium-high, medium, and low.
Previous studies have categorized the proportion of minority patients at a center as low, medium and high.14 In this study, race composition of each center was defined by the percentage of listed active patients at the center during the study period. Centers with a high proportion of Black patients (>50%) were compared against other centers.
Patient level covariates included: i.) Donor variables - age, height, weight, ethnicity, hypertension, diabetes, cause of death, creatinine, and donation after circulatory death (DCD)- these are variables used to derive the kidney donor risk index (KDRI);15 ii) Recipient variables - age, blood group, cause of end-stage renal disease, waiting time (time from dialysis start date to transplant or time from wait-listing to transplantation for preemptive transplants), insurance, education level, calculated panel reactive antibodies (cPRA) and number of HLA mismatches.16
2.6. Exclusion Criteria
We excluded recipients with unknown/missing/other race as this category was heterogenous and poorly defined. We also excluded recipients with unknown/missing education and unknown/missing insurance. In addition, recipient data from centers with no active patients on the waiting list based on data available (waitlist ID) in the STAR file were excluded. With this data missing, active waitlist size and a center’s race composition could not be determined. Finally, data from centers performing fewer than 20 transplants in the study period were also excluded, as very small volume transplant centers may bias results.17
2.7. Analytical Methods
For descriptive statistics, univariate analysis using Chi-square tests/Fisher’s Exact test for categorical variables and analysis of variance (ANOVA) or Kruskal-Wallis test for continuous variables were used, as appropriate, to compare donor and recipient variables by race. The study population was further characterized and stratified by recipient HCV serostatus with analyses performed separately for HCV-seronegative and HCV-seropositive recipients. For analyses evaluating the effect of race on receipt of an HCV-viremic kidney, we fit mixed effect multivariable logistic regression models, treating transplant center as a random effect, and adjusting for center, donor, and recipient variables. These models account for correlations between individual recipients at each transplant center with regard to receipt of an HCV-viremic kidney as patients (level 1) are nested within transplant centers (level 2).16,18
To assess whether a lag in insurance coverage in the first three years following the availability of antiviral medication in 2014 affected the relationship between race and HCV-viremic kidney utilization, sensitivity analyses were performed in which full models were run as described above with follow-up beginning January 1, 2017.
Analysis was performed using SAS version 9.4 (SAS Institute, Cary, NC). For all analyses a p-value of <0.05 was considered statistically significant.
3. Results
3.1. Study population
The study included 58,786 adult deceased donor kidney transplants. Of these recipients, 21,037 (36%) were White, 21,343 (36%) were Black, 11,734 (20%) were Hispanic and 4,672 (8%) were Asian. Among recipients, 3076 (5%) were HCV seropositive. Of these, 810 (26%) were White, 1709 (56%) were Black, 449 (15%) were Hispanic, and 108 (4%) were Asian (Table 1).
Table 1:
Recipient Characteristics (N=58,786)
| White n= 21,037 |
Black n=21,343 |
Hispanic n=11,734 |
Asian n=4,672 |
P-value | |
|---|---|---|---|---|---|
| Age, years mean ± sd | 57 ± 13 | 53 ± 12 | 51 ± 13 | 55 ± 13 | <.0001 |
| Male, n (%) | 12,743 (61) | 12,761 (60) | 7210 (61) | 2,621 (56) | <.0001 |
| Diagnosis, n (%) | |||||
| Diabetes | 5,729 (27) | 7,174 (34) | 4,713 (40) | 1,482 (32) | <.0001 |
| HTN | 3,729 (18) | 8,269 (39) | 2,645 (23) | 1,052 (23) | |
| GN | 3,701 (18) | 2,391 (11) | 1,694 (14) | 1,186 (25) | |
| Congenital | 3,318 (16) | 919 (4) | 737 (6) | 224 (5) | |
| Other | 4,560 (22) | 2,590 (12) | 1,945 (17) | 728 (16) | |
| Waiting time (months) median (IQR) | 48 (24–60) | 60 (36–84) | 60 (36–84) | 60 (36–84) | <.0001 |
| Blood Group, n (%) | |||||
| A | 9,356 (44) | 6,143 (29) | 3,803 (32) | 1,297 (28) | <.0001 |
| B | 2,043 (10) | 4,055 (19) | 1,113 (10) | 1,234 (26) | |
| AB | 1,192 (6) | 1,219 (6) | 338 (3) | 459 (10) | |
| O | 8,446 (40) | 9,926 (47) | 6,460 (55) | 1,632 (36) | |
| CPRA category n (%) | |||||
| Less than 70 | 18,642 (89) | 18,448 (86) | 10,084 (86) | 4,076 (87) | <.0001 |
| Number of HLA Mismatch median (IQR) | 4 (3–5) | 5 (4–5) | 4 (3–5) | 5 (4–5) | <.0001 |
| Education level n (%) | |||||
| Grade school or Below | 426 (2) | 544 (3) | 3,290 (28) | 511 (11) | <.0001 |
| High school | 8,237 (39) | 9,479 (44) | 5,254 (45) | 1,387 (30) | |
| College or Above | 12,374 (59) | 11,320 (53) | 3,190 (27) | 2,774 (59) | |
| Insurance, n (%) | |||||
| Medicaid | 645 (3) | 1,189 (6) | 1,428 (12) | 421 (9) | <.0001 |
| Medicare | 15,047 (72) | 17,062 (80) | 7,979 (68) | 3,076 (66) | |
| Private | 5,345 (25) | 3,092 (14) | 2,227 (20) | 1,175 (25) |
Overall, Whites and Asians were older than Blacks and Hispanics (57- and 55-years vs 53-and 51-years. For Whites, Hispanics, and Asians, diabetes was the most common primary cause for renal failure (27%, 40% and 32%, respectively). For Blacks, hypertension was the most common diagnosis (39%). Whites had a shorter average waiting time of 48 months, while all other races had an average waiting time of 60 months. Whites, Asians, and Blacks mostly had an education at the college level or above (59%, 59%, and 53%, respectively). The proportion of recipients with a grade school level of education was higher among Hispanics (28%) and Asians (11%). Medicare use was common for all races. Whites and Asians had the highest proportions of private insurance use (25%), while Hispanics had the highest proportion of Medicaid use (12%).
3.2. Characteristics of recipients of HCV-viremic kidneys
Of the transplanted kidneys, 2,613 (4%) were obtained from HCV-viremic donors. Of these, 1,598 (61%) were utilized in HCV-seronegative recipients [D+/R-]. Among these recipients, 719 (45%) were White, 627 (39%) were Black, 163 (10%) were Hispanic, and 89 (6%), were Asian. One thousand and fifteen (39%) of HCV-viremic kidneys were utilized among sero-positive recipients [D+/R+]. Of these recipients, 262 (25%) were White, 610 (60%) were Black, 118 (12%) were Hispanic, and 25 (2%) were Asian (Table 2).
Table 2:
Characteristics of HCV-seronegative (N=1598) and HCV-seropositive (N=1015) recipients of HCV viremic kidneys
| White | Black | Hispanic | Asian | |||||
|---|---|---|---|---|---|---|---|---|
|
(R−) n=719 |
(R+) n=262 |
(R−) n=627 |
(R+) n=610 |
(R−) n=163 |
(R+) n=118 |
(R−) n=89 |
(R+) n=25 |
|
| Age, years mean ± sd | 61 ± 10 | 61 ± 9 | 56 ± 11 | 60 ± 8 | 56 ± 13 | 59 ± 9 | 59 ± 10 | 62 ± 10 |
| Male, n (%) | 520 (72) | 210 (82) | 408 (65) | 469 (77) | 128 (79) | 92 (78) | 68 (76) | 19 (76) |
| Diagnosis, n (%) | ||||||||
| Diabetes | 267 (37) | 108 (41) | 274 (44) | 265 (43) | 87 (53) | 75 (64) | 41 (46) | 15 (60) |
| HTN | 133 (19) | 46 (18) | 217 (35) | 258 (42) | 27 (17) | 16 (14) | 14 (16) | 6 (24) |
| GN | 97 (13) | 29 (10) | 56 (9) | 36 (6) | 17 (10) | 12 (10) | 17 (19) | 2 (8) |
| Congenital | 114 (16) | 21 (8) | 25 (4) | 9 (1) | 17 (10) | 2 (2) | 7 (8) | 1 (4) |
| Other | 108 (15) | 58 (22) | 55 (9) | 42 (7) | 15 (9) | 13 (11) | 10 (11) | 1 (4) |
| Waiting time, months median (IQR) | 36 (12–48) | 24 (12–48) | 48 (24–60) | 36 (24–60) | 36 (24–60) | 36 (24–60) | 36 (24–60) | 36 (12–48) |
| Blood Group, n (%) | ||||||||
| A | 278 (39) | 113 (43) | 176 (28) | 149 (24) | 44 (27) | 38 (32) | 27 (30) | 8 (32) |
| B | 56 (8) | 22 (8) | 101 (16) | 123 (20) | 18 (8) | 15 (13) | 24 (27) | 8 (32) |
| AB | 30 (4) | 5 (2) | 28 (4) | 32 (5) | 4 (2) | 1 (1) | 4 (4) | 1 (4) |
| O | 355 (49) | 122 (47) | 322 (51) | 306 (50) | 102 (63) | 64 (54) | 34 (38) | 8 (32) |
| CPRA category n (%) | ||||||||
| Less than 70 | 694 (97) | 246 (94) | 603 (96) | 579 (95) | 158 (97) | 115 (97) | 87 (98) | 25 (100) |
| No of HLA Mismatch median (IQR) | 4 (4–5) | 4 (4–5) | 5 (4–5) | 5 (4–6) | 5 (4–5) | 5 (4–5) | 5 (4–5) | 5 (4–5) |
| Education level n (%) | ||||||||
| Grade School or Below | 6 (1) | 6 (2) | 15 (2) | 22 (4) | 30 (18) | 20 (17) | 7 (8) | 3 (12) |
| High School | 240 (33) | 123 (47) | 246 (39) | 331 (54) | 78 (48) | 71 (60) | 22 (25) | 11 (44) |
| College or Above | 473 (66) | 133 (51) | 366 (58) | 257 (42) | 55 (34) | 27 (23) | 60 (67) | 11 (44) |
| Insurance, n (%) | ||||||||
| Medicaid | 14 (2) | 10 (4) | 40 (6) | 54 (9) | 15 (9) | 17 (14) | 6 (7) | 3 (12) |
| Medicare | 492 (68) | 173 (66) | 466 (74) | 410 (67) | 109 (67) | 78 (66) | 62 (70) | 18 (72) |
| Private | 213 (30) | 79 (30) | 121 (19) | 146 (24) | 39 (24) | 23 (19) | 21 (24) | 4 (16) |
Utilization of HCV-viremic kidneys increased among the HCV-seronegative over time while there was a concomitant decrease among the sero-positive as shown in (Figure 1).
Figure 1.
Utilization of HCV-viremic kidneys among seronegative and seropositive recipients over time
3.3. Center Characteristics
The deceased donor kidney transplants were performed at 191 transplant centers. The median center active waitlist size was 276 patients (IQR 151–496). Thirty-three centers (17%) had an active waitlist with >50% Black patients. These 33 centers performed 36% of all transplants among Black recipients. The median waitlist size at centers with a race composition >50% Black was 390 patients (IQR 191–653) vs. 254 patients (IQR 147–469) at other centers. The average proportion of Black waitlisted patients at centers with a proportion >50% Black was 70% (Supplemental Table 1). One hundred transplant centers (52%) were categorized as low-volume, 45 centers (24%) medium-volume, 29 centers (15%) medium-high volume and 17 centers (9%) high-volume. Sixty-one centers (32%) were categorized as low-risk, 40 (21%) medium-risk, 41 (22%), medium-high risk and 49 (26%) were high-risk (Supplemental Table 2). Collectively, low-risk centers utilized 140 (5%) HCV-viremic donors, medium-risk centers utilized 284 (11%), medium-high risk centers utilized 601 (23%) and high-risk centers utilized 1588 (61%)(Supplemental Table 3).
3.4. Donor Characteristics (HCV-Viremic Kidneys)
Donors of HCV-viremic kidneys were young, with a mean age of 38 years. Donors were even younger for kidneys utilized in HCV-seropositive patients, with a mean age of 33 years. Hypertension was a diagnosis in 14% of viremic donors, while diabetes was a diagnosis in 2%. DCD constituted 15%. The median terminal creatinine was 0.9 (IQR 0.7–1.31). There were no clinically significant racial differences in these characteristics.
3.5. Factors associated with receipt of HCV-viremic kidney among HCV-seronegative patients
As shown in (Table 3), the initial multivariable model (Model 1) adjusted for donor, recipient and center variables (but not including waiting time and educational level) showed Blacks, Hispanics and Asians were less likely than Whites to receive HCV-viremic kidneys (Black OR 0.69, 95% CI [0.60,0.80]); Hispanic (OR 0.63, 95% CI [0.51,0.79]); Asian (OR 0.69, 95% CI [0.53,0.90]). After adjusting for waiting time (Model 2), where a shorter waiting time was associated with increased odds of receiving HCV-viremic kidneys (OR 0.65, 95% CI [0.63,0.67]), the previous effect identified for Black race was attenuated (OR 0.96, 95% CI [0.83,1.12]). After adjusting for educational level in the final model (Model 3), the previous effect identified for Hispanics and Asians was attenuated (Hispanic OR 0.89, 95% CI [0.70,1.11]): Asian OR 0.76, 95% CI [0.58,1.01]). In the final model, the following recipient factors were associated with receipt of HCV-viremic kidneys: male gender, increased age, decreased waiting time, increasing educational level, blood group, cause of kidney failure, lower cPRA, center volume and center risk profile.
Table 3:
Association between Race and Receipt of HCV-viremic kidneys among HCV-seronegative recipients [D+/R−]*
| Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|
| Odds Ratio (95% CI) | |||||
| Race: Black vs White | 0.69 (0.60–0.80) | 0.96 (0.83–1.12) | 0.96 (0.83–1.12) | ||
| Race: Hispanic vs White | 0.63 (0.51–0.79) | 0.79 (0.63–0.98) | 0.89 (0.70–1.11) | ||
| Race: Asian vs White | 0.69 (0.53–0.90) | 0.75 (0.57–0.99) | 0.76 (0.58–1.01) | ||
| Gender: Male vs Female | 1.25 (1.10–1.42) | 1.28 (1.12–1.47) | 1.29 (1.13–1.47) | ||
| Age (10 years)† | 1.68 (1.59–1.77) | 1.57 (1.48–1.66) | 1.57 (1.48–1.66) | ||
| Waiting time (years) | - | 0.65 (0.63–0.67) | 0.65 (0.63–0.67) | ||
| Education Level: Grade School or Below vs High School | - | - | 0.67 (0.49–0.92) | ||
| Education Level: College or Above vs High School | - | - | 1.16 (1.02–1.31) | ||
| Blood group: A vs O | 0.64 (0.56–0.73) | 0.41 (0.36–0.47) | 0.41 (0.36–0.47) | ||
| Blood group: B vs O | 0.58 (0.49–0.70) | 0.46 (0.38–0.56) | 0.46 (0.38–0.56) | ||
| Blood group: AB vs O | 0.38 (0.29–0.51) | 0.16 (0.12–0.22) | 0.16 (0.12–0.22) | ||
| Cause of kidney failure: Hypertension vs Diabetes | 0.64 (0.55–0.74) | 0.69 (0.59–0.80) | 0.69 (0.59–0.80) | ||
| Cause of kidney failure: Glomerulonephritis vs Diabetes | 0.67 (0.56–0.82) | 0.65 (0.53–0.79) | 0.64 (0.53–0.78) | ||
| Cause of kidney failure: Congenital vs Diabetes | 0.88 (0.71–1.1) | 0.94 (0.76–1.16) | 0.93 (0.75–1.15) | ||
| Cause of kidney failure: Other vs Diabetes | 0.59 (0.49–0.72) | 0.61 (0.50–0.75) | 0.60 (0.49–0.74) | ||
| cPRA: <70 vs >70 | 4.76 (3.53–6.43) | 8.74 (6.4–12.0) | 8.69 (6.32–11.94) | ||
| Medical Insurance: Medicare vs Medicaid | 0.95 (0.72–1.24) | 1.17 (0.88–1.55) | 1.11 (0.83–1.47) | ||
| Medical Insurance: Private vs Medicaid | 1.50 (1.12–1.99) | 1.11 (0.82–1.49) | 1.03 (0.76–1.39) | ||
| Center % Black Race(>50%): Yes vs No | 2.55 (1.06–6.03) | 2.21 (0.91–5.36) | 2.23 (0.92–5.40) | ||
| Center Transplant Volume: Medium vs Low | 1.43 (0.60–3.40) | 1.40 (0.57–3.42) | 1.39 (0.57–3.40) | ||
| Center Transplant Volume: Medium-High vs Low | 2.44 (0.95–6.23) | 2.42 (0.92–6.37) | 2.40 (0.91–6.31) | ||
| Center Transplant Volume: High vs Low | 4.67 (1.51–14.50) | 4.80 (1.50–15.38) | 4.75 (1.48–15.23) | ||
| Center Utilization of PHS kidneys: Medium vs Low | 1.59 (0.55–4.57) | 1.62 (0.55–4.80) | 1.62 (0.55–4.80) | ||
| Center Utilization of PHS kidneys : Medium-High vs Low | 4.36 (1.58–11.97) | 4.33 (1.53–12.26) | 4.36 (1.54–12.34 | ||
| Center Utilization of PHS kidneys : High vs Low | 24.44 (9.44–63.29) | 27.1 (10.2–72.3) | 27.3 (10.3–72.8) | ||
Recipient factors associated with receiving HCV-viremic kidneys among seronegative patients are shown.
Model 1: Adjusted for donor variables: age, height, weight, ethnicity, hypertension, diabetes, cause of death, creatinine, and non-heart beating donor (DCD) and number of HLA mismatches.
Model 2: Adjusted for variables from Model1 and waiting time
Model 3: Adjusted for variables from Model 2 and education level
Abbreviations: CI: confidence interval; cPRA: calculated panel reactive antibodies; HCV: Hepatitis C; PHS: public health service; HLA: Human leukocyte antigen
Unit for age is 10 years i.e., odds increase per 10-year increase in age.
3.6. Factors associated with receipt of HCV-viremic kidney among HCV-seropositive patients
We found no statistically significant racial differences on the individual level after adjusting for center and donor variables and recipient variables excluding waiting time and education. In the final model (Table 4), factors associated with receipt of HCV-viremic kidneys included: age; waiting time, educational level, blood group; cause of kidney failure, cPRA, insurance status, centers with a proportion >50% Black, and center risk profile.
Table 4:
Association between Race and Receipt of HCV-viremic kidneys among HCV-seropositive recipients [D+/R+]*
| Variable | OR (95% CI) |
|---|---|
| Race: Black vs White | 0.90 (0.69–1.18) |
| Race: Hispanic vs White | 0.86 (0.59–1.25) |
| Race: Asian vs White | 0.66 (0.35–1.23) |
| Gender: Male vs Female | 1.05 (0.82–1.34) |
| Age† (years) | 1.40 (1.25–1.58) |
| Waiting time (years) | 0.78 (0.75–0.81) |
| Education Level: Grade School or Below vs High School | 0.63 (0.40–0.99) |
| Education Level: College or Above vs High School | 0.84 (0.68–1.04) |
| Blood group: A vs O | 0.68 (0.54–0.86) |
| Blood group: B vs O | 1.05 (0.78–1.41) |
| Blood group: AB vs O | 0.45 (0.27–0.73) |
| Cause of kidney failure: Hypertension vs Diabetes | 0.81 (0.64–1.03) |
| Cause of kidney failure: Glomerulonephritis vs Diabetes | 0.62 (0.43–0.90) |
| Cause of kidney failure: Congenital vs Diabetes | 0.58 (0.33–0.99) |
| Cause of kidney failure: Other vs Diabetes | 0.64 (0.46–0.89) |
| cPRA: <70 vs >70 | 3.1 (2.10–4.64) |
| Medical Insurance: Medicare vs Medicaid | 0.61 (0.41–0.91) |
| Medical Insurance: Private vs Medicaid | 0.80 (0.51–1.24) |
| Center % Black Race (>50%): Yes vs No | 1.71 (1.20–2.45) |
| Center Transplant Volume: Medium vs Low | 1.13 (0.77–1.68) |
| Center Transplant Volume: Medium-High vs Low | 1.09 (0.73–1.63) |
| Center Transplant Volume: High vs Low | 1.24 (0.79–1.96) |
| Center Utilization of PHS kidneys: Medium vs Low | 2.30 (1.41–3.75) |
| Center Utilization of PHS kidneys: Medium-High vs Low | 3.04 (1.88–4.90) |
| Center Utilization of PHS kidneys: High vs Low | 3.83 (2.43–6.05) |
Recipient factors associated with receiving HCV-viremic kidneys among HCV-seropositive recipients are shown (final model).
The model is also adjusted for donor variables: age, height, weight, ethnicity, hypertension, diabetes, cause of death, creatinine and non-heart beating donor (DCD) and number of HLA mismatches.
Unit for age is 10 years i.e., odds increase per 10-year increase in age.
3.7. Sensitivity Analysis:
To assess whether a lag in insurance coverage in the first three years following the initiation of HCV-viremic kidney utilization in 2014 affected the relationship between race and utilization, sensitivity analyses were performed in which full models were run as described above with follow-up beginning January 1, 2017 (Supplemental Table 4). No large effect changes were found with exclusion of 2014–2016 follow-up; thus, sensitivity analyses were found to support using complete data from 2014 onwards in primary analyses.
4. Discussion
In the setting of increasing national utilization of HCV-viremic kidneys,19 we examined the association between recipient race and transplantation with HCV-viremic kidneys.
Compared to all deceased donor kidneys in general, HCV-viremic kidneys were distributed differently among races. Whites received a larger proportion among seronegative and Blacks the larger share among those seropositive, while Hispanics and Asians received small proportions among both seronegative and those seropositive. However, adjusted analysis revealed no independent effect of recipient race in creating the differences observed. Rather, we discovered that among HCV-seronegative recipients, potentially contributory factors driving the disparity included waiting-time and education while among the seropositive transplant center practices at centers with larger populations of Black waitlisted patients likely play an important role.
Priority for kidney transplantation accounts for waiting time based on the kidney allocation score (KAS)20 thus waitlisted individuals with longer waiting time have greater priority for transplantation. Before we adjusted for waiting time, Blacks were observed to be less likely than Whites to receive HCV-viremic kidneys. Given Blacks have longer waiting times on average compared to Whites, our study suggests that waiting time is a potential key factor mitigating the difference between Blacks and Whites when it comes to receiving HCV-viremic kidneys among seronegative recipients. This longer waiting time for Blacks is explained by a well-documented delayed access to kidney listing.21
The association between waiting time and receipt of HCV-viremic kidneys suggests that seronegative patients with longer waiting times are either not considered candidates by centers or have not given consent when an HCV-viremic kidney is offered. Transplant centers may deliberately make this choice, presuming that a non- HCV-viremic offer will be available shortly because these individuals have higher KAS scores. Also, wait-listing, evaluation and re-evaluation processes vary by center. It is possible many transplant centers have not approached significant proportions of already waitlisted patients as to the possibility of receiving HCV-viremic kidneys. Individual patient choices are important too. HCV-seronegative patients with longer waiting times may be willing to wait a little longer if it means foregoing the risk of contracting HCV.
A substantial number of HCV-viremic kidneys are still discarded.6 Our study suggests that a lot more can be done by transplant centers to encourage the use of HCV-viremic kidneys, particularly among waitlisted seronegative patients with long waiting times. Any barriers to greater use should be comprehensively studied. Such an approach will likely lead to a greater number of Black recipients of HCV-viremic kidneys among seronegative patients. The HCV-viremic kidneys utilized in this study were notably from younger donors with a median creatinine of 1, suggesting that individuals with long waiting times who accept HCV-viremic kidneys do not necessarily compromise on the quality of donor allograft.
Another important factor is education. Before we adjusted for educational level, we found that among seronegative recipients, Hispanics and Asians were less likely than Whites to receive HCV-viremic kidneys. In addition, we found that Hispanics and Asians had greater proportions of recipients with a grade school level of education. This suggests that a potential reason seronegative Hispanics and Asians are less likely to receive HCV-viremic kidneys can be explained by the fact these races have larger proportions of recipients with a lower educational level. Studies have demonstrated there is an association between educational level and medical mistrust.22 Mistrust may explain the reducing likelihood for patients with a grade school education accepting an HCV-viremic kidney among seronegative recipients. Furthermore, patients with lower education may have greater difficulty understanding the nuances of kidney transplantation and thus are less inclined to make informed decisions.23
This finding differed for HCV-seropositive recipients, whereby those with a college education level were not more likely to receive an HCV-viremic kidney compared to those with a high school level of education. A clear explanation for this observation is not apparent. However, evidence suggests transplant centers adopt different approaches to education9 and how patients are counselled plays a significant role in the final choices patients make regarding accepting kidneys. Importantly, educational approaches should consider methods that are effective for reaching patients with lower levels of education. These methods should also include cultural competency and ensure that language barriers are overcome. Aggressive approaches to mitigating this problem will likely result in greater utilization of HCV-viremic kidneys among Hispanics and Asians who currently receive a very small share of these kidneys.
Among HCV-seropositive patients, our study identified that Blacks received most HCV-viremic kidneys. This is likely because among HCV-seropositive recipients, centers with larger populations of Black waitlisted patients are almost twice as likely to utilize HCV-viremic kidneys compared to other centers. Bradley et al24 examined the prevalence of HCV in 50 US states by race and found that the prevalence of HCV among non-Hispanic Blacks was twice that of non-Hispanic Whites (Prevalence ratio 2.29 95% CI 1.94–2.70).21 Thus, these centers are likely to have larger populations of HCV-seropositive patients and may target HCV-viremic kidneys specifically to these patients. Interestingly, centers with larger populations of Black patients did not utilize HCV-viremic kidneys similarly among seronegative patients. Given such centers already have an infrastructure in place, successfully utilizing HCV-viremic kidneys among the seropositive, it stands to reason that the same strategies employed can be broadened to include seronegative patients.
Notably the utilization of HCV-viremic kidneys among the seropositive is declining over time. Our study is not designed to determine the precise cause of this however, it may be the result of early treatment of HCV prior to transplant evaluation. Clinicians should note that patients that can forgo early treatment and defer HCV treatment till post-transplant may be at an advantage as receiving HCV-viremic kidneys is associated with a shorter waiting time.
It is important to discuss the role of insurance and its association with the receipt of HCV-viremic kidneys. Direct acting antiviral medications are expensive, and utilization requires insurance approval.25 Thus, access to HCV-viremic kidneys is largely dependent the specific insurance carried by the individual. This can lead to or exacerbate racial disparity given HCV-infected individuals are more likely to be underinsured26 and among seronegative waitlisted patients those with private insurance are more likely to be White. Importantly, options for insurance coverage of HCV-viremic kidneys have increased since 2017.27 In our study, we included data from the year 2014 – the year direct acting antivirals entered in the market27 – and adjusted for insurance in all models. We then performed a sensitivity analysis using data from 2017. In final models, for both seronegative and seropositive recipients, we found that insurance had no significant effect, and no specific insurance was associated with greater receipt of HCV-viremic kidneys. This finding is likely the result of Medicare and Medicaid expanding in the year 2015 to provide coverage for direct acting antiviral medication.27
Our study identified other factors associated with the receipt of HCV-viremic kidneys which were not clearly related to how HCV-viremic kidneys are currently distributed among different racial groups but remain of interest. Among HCV-seronegative recipients, men were more likely than women to receive HCV-viremic kidneys. The SRTR annual data report28 identified rates of deceased donor kidney transplantation as similar between men and women. Thus, there is a unique aspect to the utilization of HCV-viremic kidneys among the seronegative that makes it more likely to be among men. Supporting this reasoning is the lack of a gender difference among seropositive recipients. Further work to understand the underlying mechanisms of this disparity is necessary.
We found that increasing age was associated with receiving HCV-viremic kidneys for both seronegative and seropositive patients. Older patients may feel a greater need for imminent transplantation given the awareness of health deterioration and delisting due to age while waiting for a kidney transplant9,29 and aggressively pursue HCV-viremic kidneys.
We identified multiple factors associated with receipt of HCV-viremic kidneys that could be attributed to how centers select patients for HCV-viremic kidneys including blood group, cause of kidney failure and cPRA. Initial trials among the HCV-seronegative excluded patients with a prior history of solid organ transplant.4 Centers may still be avoiding re-transplant patients and thus those with a high cPRA when they select candidates for potential HCV-viremic transplantation. In addition, centers may actively be recruiting patients they expect to deteriorate while on the waitlist e.g., those with diabetes30 and those with lower opportunities to receive a non-HCV viremic kidney transplant e.g., blood group O.31
Unsurprisingly, increasing center utilization of PHS increased risk kidneys was associated with greater receipt of HCV-viremic kidneys. Among HCV-seronegative recipients, high PHS utilizers were 27 times more likely to use HCV-viremic kidneys compared to low PHS utilizers. Among HCV-seropositive recipients, this difference was 4 times larger. Low PHS utilizers can certainly make much greater use of HCV viremic kidneys. Interestingly, center volume had a less profound effect on utilization of HCV-viremic kidneys. Among HCV-seropositive recipients, no effect was observed. Among seronegative recipients a difference was only notable between high volume and low volume centers.
Our study has some limitations. We used the race category identified in the UNOS STAR file which may differ from self-reported race, especially for non-White race/ethnicity.32 With regard to seropositive recipients, a notable limitation is our inability to determine which recipients were NAT positive or had previously received HCV eradication with antiviral medication prior to wait-listing or transplantation. Treated patients are often considered ineligible for HCV-viremic kidneys. Lastly, because our study is observational and retrospective, selection, information, and confounding biases cannot be entirely eliminated.33 However, our analyses accounted for correlation and we adjusted for a reasonable number of donor, recipient and transplant factors to reduce confounding. We also specifically designed our analysis to capture racial differences and underlying factors a priori.
While transplantation of HCV-viremic kidneys continues to increase nationally in the United States,19 many HCV-viremic kidneys are still discarded. There exists significant opportunity to increase HCV-viremic kidney utilization among all races and this can be potentially achieved by addressing unique underlying factors that contribute to reduced utilization among different races.
Supplementary Material
Funding
This study was supported by NIH grant T32DK077662 PI MM ABECASSIS MD
Abbreviations
- ANOVA
Analysis of variance
- cPRA
Calculated panel reactive antibodies
- D+
Donor positive
- DCD
Donation after cardiac death
- EXPANDER
Exploring transplants using hepatitis-C infected kidneys for HCV-negative recipients
- GN
Glomerulonephritis
- HCV
Hepatitis C
- HIV
Human immunodeficiency virus
- HLA
Human leukocyte antigen
- HTN
Hypertension
- IQR
interquartile range
- KAS
kidney allocation score
- KDPI
Kidney donor profile index
- KDRI
Kidney donor risk index
- OPTN
Organ procurement and transplant network
- PHS
public health service
- R+
Recipient positive
- R−
Recipient negative
- STAR
Standard transplant analysis file
- THINKER
Transplanting Hepatitis C Kidneys into Negative Kidney Recipients
- US
United States
Footnotes
Supporting/Supplementary Tables Information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Declarations
Conflicts of Interest/Competing interests
The authors of this manuscript declare no conflicts of interest
Availability of data and material
The data is freely available from the OPTN/UNOS
Code availability
The authors will make the code available upon request
Ethics approval
The study was approved by the Tulane University Institutional Review Board
Consent to participate
Not applicable (retrospective analysis of de-identified data)
Consent for publication
Not applicable (retrospective analysis of de-identified data)
References
- 1.Ison MG, Nalesnik MA. An update on donor-derived disease transmission in organ transplantation. Am J Transplant Jun 2011;11(6):1123–30. doi: 10.1111/j.1600-6143.2011.03493.x [DOI] [PubMed] [Google Scholar]
- 2.Seem DL, Lee I, Umscheid CA, Kuehnert MJ. PHS guideline for reducing human immunodeficiency virus, hepatitis B virus, and hepatitis C virus transmission through organ transplantation. Public Health Rep Jul 2013;128(4):247–343. doi: 10.1177/003335491312800403 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Durand CM, Zhang W, Brown DM, et al. A prospective multicenter pilot study of HIV-positive deceased donor to HIV-positive recipient kidney transplantation: HOPE in action. Am J Transplant Jul 23 2020;doi: 10.1111/ajt.16205 [DOI] [PMC free article] [PubMed]
- 4.Durand CM, Bowring MG, Brown DM, et al. Direct-Acting Antiviral Prophylaxis in Kidney Transplantation From Hepatitis C Virus-Infected Donors to Noninfected Recipients: An Open-Label Nonrandomized Trial. Ann Intern Med Apr 17 2018;168(8):533–540. doi: 10.7326/m17-2871 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Goldberg DS, Abt PL, Blumberg EA, et al. Trial of Transplantation of HCV-Infected Kidneys into Uninfected Recipients. N Engl J Med Jun 15 2017;376(24):2394–2395. doi: 10.1056/NEJMc1705221 [DOI] [PubMed] [Google Scholar]
- 6.Chang SH, Merzkani M, Lentine KL, et al. Trends in Discard of Kidneys from Hepatitis C Viremic Donors in the United States. Clin J Am Soc Nephrol Feb 8 2021;16(2):251–261. doi: 10.2215/cjn.10960720 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wachterman MW, McCarthy EP, Marcantonio ER, Ersek M. Mistrust, misperceptions, and miscommunication: a qualitative study of preferences about kidney transplantation among African Americans. Transplant Proc Mar 2015;47(2):240–6. doi: 10.1016/j.transproceed.2015.01.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Myaskovsky L, Almario Doebler D, Posluszny DM, et al. Perceived discrimination predicts longer time to be accepted for kidney transplant. Transplantation Feb 27 2012;93(4):423–9. doi: 10.1097/TP.0b013e318241d0cd [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.McCauley M, Mussell A, Goldberg D, et al. Race, Risk, and Willingness of End-Stage Renal Disease Patients Without Hepatitis C Virus to Accept an HCV-Infected Kidney Transplant. Transplantation Apr 2018;102(4):e163–e170. doi: 10.1097/tp.0000000000002099 [DOI] [PubMed] [Google Scholar]
- 10.Howell EA, Egorova N, Balbierz A, Zeitlin J, Hebert PL. Black-white differences in severe maternal morbidity and site of care. Am J Obstet Gynecol Jan 2016;214(1):122.e1–7. doi: 10.1016/j.ajog.2015.08.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Skinner J, Chandra A, Staiger D, Lee J, McClellan M. Mortality after acute myocardial infarction in hospitals that disproportionately treat black patients. Circulation Oct 25 2005;112(17):2634–41. doi: 10.1161/circulationaha.105.543231 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.OPTN. Accessed 4/10/20, https://optn.transplant.hrsa.gov/data/request-data/data-request-instructions/
- 13.Sonnenberg EM, Cohen JB, Hsu JY, et al. Association of Kidney Transplant Center Volume With 3-Year Clinical Outcomes. Am J Kidney Dis Oct 2019;74(4):441–451. doi: 10.1053/j.ajkd.2019.02.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ross-Driscoll K, Kramer M, Lynch R, Plantinga L, Wedd J, Patzer R. Variation in Racial Disparities in Liver Transplant Outcomes Across Transplant Centers in the United States. Liver Transpl Apr 2021;27(4):558–567. doi: 10.1002/lt.25918 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rao PS, Schaubel DE, Guidinger MK, et al. A comprehensive risk quantification score for deceased donor kidneys: the kidney donor risk index. Transplantation Jul 27 2009;88(2):231–6. doi: 10.1097/TP.0b013e3181ac620b [DOI] [PubMed] [Google Scholar]
- 16.Meyers MR, Shults J, Laskin B, et al. Use of public health service increased risk kidneys in pediatric renal transplant recipients. Pediatr Transplant Aug 2019;23(5):e13405. doi: 10.1111/petr.13405 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Garonzik-Wang JM, James NT, Weatherspoon KC, et al. The aggressive phenotype: center-level patterns in the utilization of suboptimal kidneys. Am J Transplant Feb 2012;12(2):400–8. doi: 10.1111/j.1600-6143.2011.03789.x [DOI] [PubMed] [Google Scholar]
- 18.Holscher CM, Bowring MG, Haugen CE, et al. National Variation in Increased Infectious Risk Kidney Offer Acceptance. Transplantation Oct 2019;103(10):2157–2163. doi: 10.1097/tp.0000000000002631 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bowring MG, Shaffer AA, Massie AB, et al. Center-level trends in utilization of HCV-exposed donors for HCV-uninfected kidney and liver transplant recipients in the United States. Am J Transplant Aug 2019;19(8):2329–2341. doi: 10.1111/ajt.15355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.OPTN. OPTN policy 8- Allocation of kidneys Accessed 7/2020, 2020. https://optn.transplant.hrsa.gov/media/1200/optn_policies.pdf
- 21.Patzer RE, Plantinga LC, Paul S, et al. Variation in Dialysis Facility Referral for Kidney Transplantation Among Patients With End-Stage Renal Disease in Georgia. Jama Aug 11 2015;314(6):582–94. doi: 10.1001/jama.2015.8897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sutton AL, He J, Edmonds MC, Sheppard VB. Medical Mistrust in Black Breast Cancer Patients: Acknowledging the Roles of the Trustor and the Trustee. J Cancer Educ Jun 2019;34(3):600–607. doi: 10.1007/s13187-018-1347-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Taylor DM, Bradley JA, Bradley C, et al. Limited health literacy is associated with reduced access to kidney transplantation. Kidney Int May 2019;95(5):1244–1252. doi: 10.1016/j.kint.2018.12.021 [DOI] [PubMed] [Google Scholar]
- 24.Bradley H, Hall EW, Rosenthal EM, Sullivan PS, Ryerson AB, Rosenberg ES. Hepatitis C Virus Prevalence in 50 U.S. States and D.C. by Sex, Birth Cohort, and Race: 2013–2016. Hepatol Commun Mar 2020;4(3):355–370. doi: 10.1002/hep4.1457 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Henry B DRUG PRICING & CHALLENGES TO HEPATITIS C TREATMENT ACCESS. J Health Biomed Law Sep 2018;14:265–283. [PMC free article] [PubMed] [Google Scholar]
- 26.Stepanova M, Kanwal F, El-Serag HB, Younossi ZM. Insurance status and treatment candidacy of hepatitis C patients: analysis of population-based data from the United States. Hepatology Mar 2011;53(3):737–45. doi: 10.1002/hep.24131 [DOI] [PubMed] [Google Scholar]
- 27.Torabi J, Rocca JP, Ajaimy M, et al. Commercial insurance delays direct-acting antiviral treatment for hepatitis C kidney transplantation into uninfected recipients. Transpl Infect Dis Feb 2021;23(1):e13449. doi: 10.1111/tid.13449 [DOI] [PubMed] [Google Scholar]
- 28.Hart A, Smith JM, Skeans MA, et al. OPTN/SRTR 2018 Annual Data Report: Kidney. Am J Transplant Jan 2020;20 Suppl s1:20–130. doi: 10.1111/ajt.15672 [DOI] [PubMed] [Google Scholar]
- 29.Schold J, Srinivas TR, Sehgal AR, Meier-Kriesche HU. Half of kidney transplant candidates who are older than 60 years now placed on the waiting list will die before receiving a deceased-donor transplant. Clin J Am Soc Nephrol Jul 2009;4(7):1239–45. doi: 10.2215/cjn.01280209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med Dec 2 1999;341(23):1725–30. doi: 10.1056/nejm199912023412303 [DOI] [PubMed] [Google Scholar]
- 31.Glander P, Budde K, Schmidt D, et al. The ‘blood group O problem’ in kidney transplantation--time to change? Nephrol Dial Transplant Jun 2010;25(6):1998–2004. doi: 10.1093/ndt/gfp779 [DOI] [PubMed] [Google Scholar]
- 32.Boehmer U, Kressin NR, Berlowitz DR, Christiansen CL, Kazis LE, Jones JA. Self-reported vs administrative race/ethnicity data and study results. Am J Public Health Sep 2002;92(9):1471–2. doi: 10.2105/ajph.92.9.1471 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Grimes DA, Schulz KF. Bias and causal associations in observational research. Lancet Jan 19 2002;359(9302):248–52. doi: 10.1016/s0140-6736(02)07451–2 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.