Trends in Discard of Kidneys from Hepatitis C Viremic Donors in the United States

Visual Abstract

Abstract

Background and objectives

Kidneys from hepatitis C virus (HCV) viremic donors have become more commonly accepted for transplant, especially after effective direct-acting antiviral therapy became available in 2014. We examined the contemporary trend of kidney discard from donors with HCV seropositivity and viremia.

Design, setting, participants, & measurements

Data from the Organ Procurement and Transplantation Network were used to identify deceased donor kidneys recovered for transplant. The exposure was donor HCV antibody status in the first analyses, and donor HCV antibody and viremia status in the second analyses. Multilevel, multivariable logistic regression was used to assess the association of these HCV exposure measures with kidney discard, adjusted for donor characteristics. Multilevel analyses were conducted to account for similar kidney discard pattern within clusters of organ procurement organizations and regions.

Results

Among 225,479 kidneys recovered from 2005 to 2019, 5% were from HCV seropositive donors. Compared with HCV seronegative kidneys, the odds of HCV seropositive kidney discard gradually declined, from a multivariable-adjusted odds ratio (aOR) of 7.06 (95% confidence interval [95% CI], 5.65 to 8.81) in 2014, to 1.20 (95% CI, 1.02 to 1.42) in 2019. Among 82,090 kidneys with nucleic acid amplification test results in 2015–2019, 4% were from HCV viremic donors and 2% were from aviremic seropositive donors. Compared with HCV aviremic seronegative kidneys, the odds of HCV viremic kidney discard decreased from an aOR of 4.89 (95% CI, 4.03 to 5.92) in 2018, to 1.48 (95% CI, 1.22 to 1.81) in 2019. By 2018 and 2019, aviremic seropositive status was not associated with higher odds of discard (2018: aOR, 1.13; 95% CI, 0.88 to 1.45; and 2019: aOR, 0.97; 95% CI, 0.76 to 1.23).

Conclusions

Despite the decrease in kidney discard in recent years, kidneys from viremic (compared with aviremic seronegative) donors still had 48% higher odds of discard in 2019. The potential of these discarded organs to provide successful transplantation should be explored.

Introduction

Kidney transplantation is the most beneficial and cost-effective treatment for patients with ESKD (1–3). As the prevalence of ESKD in the United States has grown exponentially, combined with recognition of transplant benefit even for older patients and those with comorbidity, the need for kidney transplantation has been greatly exceeding the available kidney organs in the past few decades. As a result, kidney organ shortage continues to worsen. However, nearly 20% of these valuable deceased donor kidneys were discarded in 2016 (4).

Hepatitis C virus (HCV) infection in donors has, historically, been a major driver of organ discards (5–7). These include donors with prior HCV exposure and donors with HCV viremia as detected by nucleic acid amplification testing (NAT). Since 2014, the use of direct-acting antivirals (DAAs) for HCV treatment has resulted in a cure rate of >90% (8), which has revolutionized the practice for managing HCV infection. Furthermore, DAAs have shown to be safe and efficient in patients with CKD and kidney transplant recipients (9–12). This strategy has helped increase the acceptability of kidneys from donors with HCV infection in many transplant centers, and facilitated their acceptance by patients of CKD.

The purpose of this study is to examine the contemporary trend of discard of kidneys from donors with HCV viremia, up to 2019. Moreover, the study aims to assess whether aviremic donor status remains a risk factor for kidney discard.

Materials and Methods

Study Design and Data

We conducted a retrospective study using data on deceased donors from the Organ Procurement and Transplantation Network (OPTN). A detailed description of the OPTN data has been previously published (13,14). Briefly, the OPTN contains all national data on the candidate waiting list, organ donation and matching, and transplantation. Therefore, the study population was all kidney organs from deceased donors that were recovered for transplantation.

Exempt study approval was obtained from the Washington University School of Medicine’s Institutional Review Board (approval #202006076).

Outcome

Deceased donor kidney discard was the outcome. Kidney discard was defined as kidney recovered for transplant but not transplanted, which was ascertained by nonmissing reasons for kidney discard.

Exposure

Before 2015, HCV was only captured through donor HCV antibody status (HCV seropositive and HCV seronegative), which was the exposure for the first analyses. The exposure of the second analyses was donor HCV antibody and viremia status: viremic, i.e., NAT positive, regardless of HCV antibody status (HCV seropositive or seronegative) (15); aviremic seropositive; and aviremic seronegative.

Eligible Kidneys from Deceased Donors

From 2005 to 2019, a total of 129,695 deceased donors with at least one organ recovered for transplant were included (Figure 1). Among these donors, 24,892 kidneys were not recovered, which resulted in 234,498 kidneys recovered for transplant. Among these 234,498 recovered kidneys, 185 were from donors with missing HCV antibody status. Additionally, 8834 kidneys (3%) had missing values on at least one variable in the analyses. After excluding these kidneys, the eligible kidneys in the first analysis contained 225,479 deceased donor kidneys.

Figure 1.

Data attrition diagram: Total sample size was 225,479 kidneys for the first analyses and 82,090 kidneys for the second analyses. Aviremic refers to negative NAT, and viremic refers to positive NAT. ^†NAT result available date. HCV, hepatitis C virus; NAT, nucleic acid amplification test.

NAT results were available after March 31, 2015. Therefore, we excluded 143,389 kidneys recovered for transplant (72,122 donors) with donor HCV antibody status before that date and so missing NAT results. The eligible kidneys in the second analysis comprised 82,090 deceased donor kidneys.

Statistical Analyses

Summary statistics on donor characteristics stratified by donor HCV antibody status (HCV seropositive versus seronegative) were conducted, followed by summary statistics on donor characteristics stratified by HCV antibody and viremia status (viremic versus aviremic seropositive versus aviremic seronegative). Additionally, donor characteristics for discarded kidneys by HCV antibody and viremia status were also presented.

To account for similar kidney discard pattern within clusters of organ procurement organizations and United Network for Organ Sharing (UNOS) regions, we performed analyses using multilevel models (16), whenever possible. In the first analyses, multivariable, multilevel logistic regression was used to examine the association of donor HCV antibody status with kidney discard in each year, adjusted for donor age (<18, 18–30, 31–44, 45–59, and ≥60 years), race (White, Black, and Hispanic), sex, body mass index (<18.5, 18.5–24.9, 25–29.9, 30–34.9, and ≥35 kg/m²), diabetes (yes, no), hypertension (yes, no), kidney donor profile index score (KDPI; ≤20%, 21%–35%, 36%–84%, and ≥85%), donation after cardiac death (yes, no), cause of death (anoxia, cerebrovascular/stroke, head trauma, and central nervous system tumor), risk factors for blood-borne disease (yes, no), biopsy status (yes, no), and glomerulosclerosis (0–5, 6–10, 11–15, 16–20, and >20 glomeruli, indeterminate, and unknown). In the second analyses, multivariable, multilevel logistic regression was used to examine the association of donor HCV antibody and viremia status (viremic versus aviremic seronegative, and aviremic seropositive versus aviremic seronegative) with kidney discard in each year, adjusted for the same donor characteristics. Using multilevel multivariable logistic regression in both analyses may reduce the bias resulting from the clustering of the discard pattern and confounding donor characteristics.

In both analyses, multivariable-adjusted odds ratio (aOR) and 95% confidence interval (95% CI) for kidney discard in each year was plotted to assess the trend from 2012 to 2019 for the first analysis, and from 2015 to 2019 for the second analysis. All tests were two-sided. Statistical significance was determined by an α level of 0.05. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Supplemental Analyses

To consider the role that KDPI plays in kidney discard, we stratified the overall trend of kidney discard and viremic, aviremic seropositive, and aviremic seronegative kidney discard in percentages and numbers by KDPI score (≤20%, 21%–35%, 36%–84%, and ≥85%). Multivariable, multilevel logistic regression was also performed to examine the association of donor HCV antibody and viremia status (viremic versus aviremic seronegative, and aviremic seropositive versus aviremic seronegative) with kidney discard, stratified by KDPI score (<85%, ≥85%) in each year, adjusted for the same donor characteristics. KDPI score in the multivariable, multilevel analyses was stratified by <85% and ≥85% because of the small sample sizes of HCV viremic and aviremic seropositive kidneys in each of the three lower-scoring KDPI groups.

Furthermore, we explored the trend, geographical distribution, and reasons for viremic kidney discard, from 2018 to 2019. UNOS regions were used to characterize the pattern of geographical variations. Viremic kidney discard and the reasons for viremic kidney discard were tabulated by UNOS region and year.

Last, we compared post-transplant 1-year eGFRs between patients receiving HCV viremic kidneys and patients receiving HCV aviremic seronegative kidneys in the same KDPI group (50% to <85%, or ≥85%). For a fair comparison within the same KDPI group, we performed a 1:1 nearest neighbor propensity score matching between the two groups on the following recipient and transplant characteristics: age, race, sex, eGFR at transplant, diabetes status, and cold ischemia time. One-year eGFR was computed with the creatinine measurement closest to the 1 year using the CKD Epidemiology Collaboration equation (17) for recipients with available post-transplant creatinine measurements in the OPTN. For each KDPI group, the mean eGFR between patients receiving HCV viremic kidneys and the matched patients receiving HCV aviremic seronegative kidneys was compared by a t test.

Results

Donor Characteristics

Among a total of 225,479 deceased donor kidneys recovered for transplantation between January 1, 2005 and September 30, 2019, 5% (n=10,706) were from HCV seropositive donors (Figure 1). Summary statistics on donor characteristics by donor HCV antibody status are presented in Table 1. Compared with donor HCV seronegative kidneys, donors of HCV seropositive kidneys were less likely to be <18 or ≥60 years old; more likely to be male, White, and normal weight or overweight; less likely to have a KDPI score ≤20% or donation after cardiac death; and more likely to have cause of death as anoxia, risk factors for blood-borne disease, biopsy performed, and glomerulosclerosis category of 0–5 glomeruli.

Table 1.

Donor characteristics for kidneys recovered for transplantation from January 1, 2005 to September 30, 2019, by donor hepatitis C virus antibody status

Donor Characteristics	HCV Seronegative, n=214,773		HCV Seropositive, n=10,706		Total, n=225,479
	n	%	n	%	n	%
Age, yr
<18	23,594	11	43	0.40	23,637	10
18–30	50,470	23	2635	25	53,105	24
31–44	46,303	22	3596	34	49,899	22
45–59	66,465	31	3950	37	70,415	31
≥60	27,941	13	482	5	28,423	13
Sex
Female	86,376	40	4121	38	90,497	40
Male	128,397	60	6585	62	134,982	60
Race
White	143,962	67	8332	78	152,294	68
Black	31,897	15	1185	11	33,082	15
Hispanic	30,004	14	992	9	30,996	14
Other	8910	4	197	2	9107	4
Body mass index, kg/m2
<18.5	13,901	6	274	3	14,175	6
18.5–24.9	72,495	34	4111	38	76,606	34
25–29.9	65,277	30	3671	34	68,948	31
30–34.9	35,780	17	1653	15	37,433	17
≥35	27,320	13	997	9	28,317	13
Diabetes status
No	193,866	90	9941	93	203,807	90
Yes	20,907	10	765	7	21,672	10
Hypertension status
No	146,187	68	7603	71	153,790	68
Yes	68,586	32	3103	29	71,689	32
Kidney donor profile index, %
<25	50,527	24	121	1	50,648	22
25–49	32,982	15	1327	12	34,309	15
50–84	98,495	46	6802	64	105,297	47
≥85	32,769	15	2456	23	35,225	16
Donation after cardiac death
No	182,060	85	9436	88	191,496	85
Yes	32,713	15	1270	12	33,983	15
Cause of death
Anoxia	65,040	30	5238	49	70,278	31
Cerebrovascular/stroke	72,934	34	2876	27	75,810	34
Head trauma	75,745	35	2582	24	78,327	35
CNS tumor	1054	0.49	10	0.09	1064	0.47
Discard
No	177,543	83	6101	57	183,644	81
Yes	37,230	17	4605	43	41,835	19
Per PHS, does the donor have risk factors for blood-borne disease transmission?
No	186,938	87	4325	40	191,263	85
Yes	27,835	13	6381	60	34,216	15
Biopsy status
No	110,107	51	4397	41	114,504	51
Yes	104,666	49	6309	59	110,975	49
Glomerulosclerosis
0–5	64,192	30	4328	40	68,520	30
6–10	15,819	7	871	8	16,690	7
11–15	7700	4	346	3	8046	4
16–20	4806	2	227	2	5033	2
≥20	11,135	5	467	4	11,602	5
Indeterminate	1176	0.55	81	0.76	1257	0.56
Unknown	109,945	51	4386	41	114,331	51

HCV, hepatitis C virus; CNS, central nervous system; PHS, public health service.

Among a total of 82,090 kidneys recovered for transplantation between March 31, 2015 and September 30, 2019 with HCV antibody and NAT information (Figure 1), 4% (n=3640) were viremic and 2% (n=1984) were aviremic seropositive. Summary statistics on donor characteristics by HCV antibody and viremia status (viremic, aviremic seropositive, and aviremic seronegative) are presented in Table 2. Compared with aviremic seronegative kidneys, donors for viremic kidneys were less likely to be <18 or ≥60 years of age; more likely to be male, White, and normal weight or overweight; less likely to have diabetes, hypertension, a KDPI score ≤20% or donation after cardiac death; and more likely to have cause of death as anoxia, risk factors for blood-borne disease, and glomerulosclerosis category of 0–5 glomeruli. Donors for aviremic seropositive kidneys were less likely to be <18 years of age; more likely to be White and overweight; less likely to have a KDPI score <25% or donation after cardiac death; and more likely to have cause of death as anoxia, risk factors for blood-borne disease, biopsy performed, and glomerulosclerosis category of 0–5 glomeruli.

Table 2.

Donor characteristics for kidneys recovered for transplantation from March 31, 2015 to September 30, 2019, by donor hepatitis C virus antibody and viremia status

Donor Characteristics	Aviremic Seropositive, n=1984		Viremic, n=3640		Aviremic Seropositive, n=76,466		Total, n=82,090
	n	%	N	%	n	%	n	%
Age, yr
<18	6	0.30	8	0.22	7522	10	7536	9
18–30	470	24	1293	36	17,978	24	19,741	24
31–44	737	37	1537	42	18,082	24	20,356	25
45–59	619	31	699	19	23,663	31	24,981	30
≥60	152	8	103	3	9221	12	9476	12
Sex
Female	937	47	1282	35	30,118	39	32,337	39
Male	1047	53	2358	65	46,348	61	49,753	61
Race
White	1675	84	3001	82	50,473	66	55,149	67
Black	108	5	239	7	11,576	15	11,923	15
Hispanic	161	8	320	9	10,916	14	11,397	14
Other	40	2	80	2	3501	5	3621	4
Body mass index, kg/m2
<18.5	46	2	77	2	4584	6	4707	6
18.5–24.9	589	30	1495	41	24,119	32	26,203	32
25–29.9	680	34	1276	35	22,837	30	24,793	30
30–34.9	395	20	507	14	13,711	18	14,613	18
≥35	274	14	285	8	11,215	15	11,774	14
Diabetes status
No	1804	91	3501	96	68,706	90	74,011	90
Yes	180	9	139	4	7760	10	8079	10
Hypertension status
No	1361	69	3021	83	51,484	67	55,866	68
Yes	623	31	619	17	24,982	33	26,224	32
Kidney donor profile index (%)
<25	14	0.71	53	1	17,363	23	17,430	21
25–49	222	11	634	17	12,164	16	13,020	16
50–84	1271	64	2569	71	35,652	47	39,492	48
≥85	477	24	384	11	11,287	15	12,148	15
Donation after cardiac death
No	1676	84	3087	85	61,032	80	65,795	80
Yes	308	16	553	15	15,434	20	16,295	20
Cause of death
Anoxia	1344	68	2504	69	31,122	41	34,970	43
Cerebrovascular/stroke	324	16	465	13	21,226	28	22,015	27
Head trauma	314	16	665	18	23,821	31	24,800	30
CNS tumor	2	0.10	6	0.16	297	0.39	305	0.37
Discard
No	1360	69	2322	64	62,519	82	66,201	81
Yes	624	31	1318	36	13,947	18	15,889	19
Per PHS, does the donor have risk factors for blood-borne disease transmission?
No	606	31	646	18	61,217	80	62,469	76
Yes	1378	69	2994	82	15,249	20	19,621	24
Biopsy status
No	746	38	1642	45	36,233	47	38,621	47
Yes	1238	62	1998	55	40,233	53	43,469	53
Glomerulosclerosis
0–5	794	40	1471	40	24,478	32	26,743	33
6–10	198	10	255	7	6339	8	6792	8
11–15	69	3	94	3	3039	4	3202	4
16–20	51	3	64	2	1921	3	2036	2
≥20	119	6	108	3	4322	6	4549	6
Indeterminate	10	0.50	9	0.25	225	0.29	244	0.30
Unknown	743	37	1639	45	36,142	47	38,524	47

Aviremic, donor nucleic acid amplification test negative; viremic, donor nucleic acid amplification test positive; CNS, central nervous system; PHS, public health service.

Summary statistics on donor characteristics for discarded kidneys by HCV antibody and viremia status are presented in Supplemental Table 1. Among 15,899 discarded kidneys, 7% were viremic and 3% were aviremic seropositive. Compared with aviremic seronegative discarded kidneys, donors for discarded viremic kidneys were less likely to be ≥60 years of age; more likely to be male, White, and normal weight or overweight; less likely to have diabetes, hypertension, a KDPI score ≤20%, or biopsy performed; and more likely to have cause of death as anoxia, risk factors for blood-borne disease, and glomerulosclerosis category of 0–5 glomeruli. Donors for aviremic seropositive discarded kidneys were less likely to be <18 or ≥60 years of age; more likely to be White and normal weight or overweight; less likely to have diabetes, hypertension, a KDPI score <25%, or biopsy performed; and more likely to have cause of death as anoxia, risk factors for blood-borne disease, and glomerulosclerosis category of 0–5 glomeruli.

Trend in Multivariable-Adjusted Odds Ratios for Donor Hepatitis C Virus Seropositive Kidney Discards, Relative to Donor Hepatitis C Virus Seronegative Kidneys

Compared with HCV seronegative kidneys, donor HCV seropositive kidneys consistently demonstrated an elevated likelihood of discard (Figure 2). In the pre-DAA era (before 2014), compared with HCV seronegative kidneys, donor HCV seropositive kidney discard reached its peak in 2013 (aOR, 7.34; 95% CI, 5.87 to 9.17). In the post-DAA era (2014 and onward), compared with HCV seronegative kidneys, the odds of donor HCV seropositive discard gradually declined starting from 2014 (aOR, 7.06; 95% CI, 5.65 to 8.81), continued to decline, and reached its lowest odds in 2019 (aOR, 1.20; 95% CI, 1.02 to 1.42).

Figure 2.

Multivariable-adjusted odds ratio for donor HCV seropositive (compared with donor HCV seronegative) kidney discard by year demonstrate a decreasing trend in the post-DAA era. Multivariable, multilevel logistic regression was used to examine the association of donor HCV antibody status with kidney discard in each year, adjusted for donor age (<18, 18–30, 31–44, 45–59, and ≥60 years), race (White, Black, and Hispanic), sex, body mass index (<18.5, 18.5–24.9, 25–29.9, 30–34.9, and ≥35 kg/m²), diabetes (yes, no), hypertension (yes, no), kidney donor profile index score (≤20%, 21%–35%, 36%–84%, and ≥85%), donation after cardiac death (yes, no), cause of death (anoxia, cerebrovascular/stroke, head trauma, and central nervous system tumor), risk factors for blood-borne disease (yes, no), biopsy status (yes, no), and glomerulosclerosis (0–5, 6–10, 11–15, 16–20, and >20 glomeruli, indeterminate, and unknown). ^†2019 data are from January 1 to September 30. DAA, direct-acting antiviral.

Trend in Multivariable-Adjusted Odds Ratios for Donor Hepatitis C Virus Viremic and Aviremic Seropositive Kidney Discards, Relative to Aviremic Seronegative Kidneys

Compared with aviremic seronegative kidneys, viremic kidneys consistently demonstrated an elevated likelihood of discard for all years (Figure 3). The odds of viremic kidney discard increased from 2015 to 2017 (2015: aOR, 5.27; 95% CI, 4.07 to 6.84; and 2017: aOR, 7.09; 95% CI, 5.71 to 8.80), and started to decrease in 2018 (aOR, 4.89; 95% CI, 4.03 to 5.92). By 2019, the odds of viremic kidney discard reached its minimum (aOR, 1.48; 95% CI, 1.22 to 1.81), but was still higher than aviremic seronegative kidneys, with a statistically significant difference.

Figure 3.

Multivariable-adjusted odds ratio for donor HCV viremic and aviremic seropositive kidney discard by year demonstrate a decreasing trend in recent years compared with aviremic seronegative kidneys. Multivariable, multilevel logistic regression was used to examine the association of donor HCV antibody and viremia status with kidney discard in each year, adjusted for donor age (<18, 18–30, 31–44, 45–59, and ≥60 years), race (White, Black, and Hispanic), sex, body mass index (<18.5, 18.5–24.9, 25–29.9, 30–34.9, and ≥35 kg/m²), diabetes (yes, no), hypertension (yes, no), kidney donor profile index score (≤20%, 21%–35%, 36%–84%, and ≥85%), donation after cardiac death (yes, no), cause of death (anoxia, cerebrovascular/stroke, head trauma, and CNS tumor), risk factors for blood-borne disease (yes, no), biopsy status (yes, no), and glomerulosclerosis (0–5, 6–10, 11–15, 16–20, and >20 glomeruli, indeterminate, and unknown). Squares indicate viremic donors and diamonds indicate aviremic seropositive donors. ^†2015 data are from March 31 to December 31. ^‡2019 data are from January 1 to September 30.

Compared with aviremic seronegative kidneys, the odds of HCV aviremic seropositive kidney discards increased from 2015 to 2016 (2015: aOR, 5.96; 95% CI, 4.12 to 8.63; and 2016: aOR, 5.52; 95% CI, 4.13 to 7.37), but started to decrease in 2017 (aOR, 2.33; 95% CI, 1.78 to 3.06). By 2018 and 2019, aviremic seropositive kidney discards were not statistically significantly different from aviremic seronegative kidney discards (2018: aOR, 1.13; 95% CI, 0.88 to 1.45; and 2019: aOR, 0.97; 95% CI, 0.76 to 1.23).

Trend in Discard of Viremic, Aviremic Seropositive, and Aviremic Seronegative Kidneys, Stratified by Kidney Donor Profile Index Score

Trend for viremic, aviremic seropositive, and aviremic seronegative kidney discard, stratified by KDPI scores ≤20%, 21%–35%, 36%–84%, and ≥85%, are presented in Supplemental Figures 1 and 2. Overall, the proportions of kidney discard increased by KDPI score, with a similar trend in the past 5 years (Supplemental Figure 1). For viremic and aviremic seropositive kidneys, there was a significantly lower discard rate of kidneys that were recovered for transplant from donors with KDPI scores of <85 and ≥85% (Supplemental Figure 2), over time. Similarly, there was a decrease in the number of discarded viremic and aviremic seropositive kidneys for donors with KDPI scores <85%, but an increase in discard for donors with KDPI scores ≥85%.

Viremic Kidney Discard: Geographical Variations, Trend, and Discard Reasons

Significant differences in the use of viremic kidneys were noted across UNOS regions (see Table 3). Regions with the greatest percentage of viremic kidney discard (i.e., number of viremic kidney discarded/number of viremic kidney recovered×100%) during 2015–2019 included region 5 (50%), region 8 (47%), region 4 (45%), and region 6 (44%). In terms of the trend across UNOS regions (see Figure 4), despite a generally decreasing trend, region 5 constantly had high percentages of viremic kidney discard, whereas region 1 constantly had low percentages of viremic kidney discard.

Table 3.

Viremic kidney discard by United Network for Organ Sharing region and year

Year	United Network for Organ Sharing Region											Total
	1	2	3	4	5	6	7	8	9	10	11
Viremic kidney recovered
2015a	46	75	53	30	47	8	12	24	14	76	36	421
2016	60	176	84	34	54	10	14	35	25	71	74	637
2017	48	150	82	25	44	17	26	35	36	72	117	652
2018	60	154	158	58	70	18	40	57	30	119	140	904
2019b	54	210	172	68	106	24	54	39	24	145	130	1026
2015–2019	268	765	549	215	321	77	146	190	129	483	497	3640
Viremic kidney discarded
2015a	18	31	21	15	25	3	6	7	6	33	10	175
2016	15	60	29	19	28	8	3	15	10	23	38	248
2017	12	52	32	14	20	9	9	22	17	42	51	280
2018	15	55	53	30	38	5	17	33	10	45	49	350
2019b	11	39	54	18	49	9	14	13	5	31	22	265
2015–2019	71	237	189	96	160	34	49	90	48	174	170	1318
Viremic kidney discarded, %
2015	39	41	40	50	53	38	50	29	43	43	28	42
2016	25	34	35	56	52	80	21	43	40	32	51	39
2017	25	35	39	56	45	53	35	63	47	58	44	43
2018	25	36	34	52	54	28	43	58	33	38	35	39
2019	20	19	31	26	46	38	26	33	21	21	17	26
2015–2019	26	31	34	45	50	44	34	47	37	36	34	36

^a2015 data are from March 31 to December 31.

^b2019 data are from January 1 to September 30.

Figure 4.

Percentages of viremic kidney discard by United Network for Organ Sharing region (1–11) and year (2015–2019) show a generally decreasing trend.

The main reason for HCV viremic kidney discard (n=1318) was no recipient located–list exhausted (782 kidneys, 59%), followed by biopsy findings (161 kidneys, 12%), other reasons (143 kidneys, 11%), and positive for hepatitis (50 kidneys, 4%). The trend for HCV viremic kidney discard reasons by UNOS region (regions 1–11) and year (2005–2019) is presented in Supplemental Figure 3. For the five regions that had highest viremic kidney discards (regions 2, 3, 5, 10, and 11), no recipient located–list exhausted was the main reason for discard in the respective year (or in all years), usually followed by biopsy findings in 2017 or 2018.

Comparison of 1-Year Post-Transplant eGFR between Patients Receiving Hepatitis C Virus Viremic Kidneys and Patients Receiving Hepatitis C Virus Aviremic Seronegative Kidneys

In donors with a KDPI score of 50% to <85%, 960 patients receiving viremic kidneys were matched to 961 patients receiving aviremic seronegative kidneys; in donors with a KDPI score of ≥85%, 78 patients receiving viremic kidneys were matched to 78 patients receiving aviremic seronegative kidneys. After excluding patients without 1-year eGFR data, there remained a total of 653 patients in the KDPI score 50% to <85% group, and a total of 41 patients in the KDPI score ≥85% group. The mean 1-year eGFR was statistically different between the two groups, i.e., patients receiving HCV viremic kidneys and patients receiving HCV aviremic seronegative kidneys (P<0.001 for the KDPI score 50% to <85% group, and P=0.03 for the KDPI score ≥85% group) (see Supplemental Figure 4). The graphs suggest higher 1-year eGFR for patients receiving HCV viremic kidneys compared with patients receiving aviremic seronegative kidneys.

Discussion

Our study examined the likelihood of discard of kidneys from HCV seropositive donors and demonstrated a decrease from 7.3 times the odds of discard in 2013 to 1.2 times in 2019, when compared with kidneys from HCV seronegative donors. To distinguish whether the decline in discard was truly in kidneys from viremic donors, we included the viremia status and found a similar trend of decrease in the adjusted odds of kidney discard from viremic donors, from 7.09 in 2017 to 1.48 in 2019, compared with kidneys from aviremic seronegative donors.

Transplantation of kidneys from aviremic seropositive donors carries a very minimal risk of HCV transmission (18–20). A series from 2018 reported that among 32 HCV seronegative patients who received aviremic seropositive kidneys, no patients developed HCV viremia or required DAA treatment (20). With increasing confidence about the safety of these organs, we identified a major change in practice. Unlike the period between 2015 and 2017, aviremic seropositive kidneys did not carry a higher risk of discard in 2018 and 2019 compared with aviremic seronegative kidneys. As more people are receiving treatment for HCV, we expect to see more HCV aviremic seropositive donors in the future, and these data suggest that these kidneys are, appropriately, being widely accepted.

Few studies explored the use of DAA prophylaxis or treatment in HCV seronegative patients who received viremic kidneys and showed successful HCV treatment and kidney transplant outcomes (21–23). HCV seronegative recipients who received viremic kidneys demonstrated no difference in allograft function at 1 year compared with those who received aviremic kidneys (23). In the short term, there was no difference in the cumulative prevalence of acute rejection or delayed graft function (23). Molnar et al. evaluated the safety of transplanting kidneys from viremic donors in aviremic seronegative recipients. All 53 recipients were found to be viremic after transplant; however, after completing antiviral treatment, all were HCV RNA negative and achieved a 12-week sustained virologic response (24). Despite four recipients with acute rejection, they concluded that kidney transplantation from viremic donors to aviremic seronegative recipients should be considered in all eligible patients (24). Another study investigating recipients of organs (including ten kidneys and one kidney-pancreas) from HCV-infected donors demonstrated that short-course DAA and ezetimibe treatments can prevent the establishment of chronic HCV infection in the recipient (25). A more recent study, and the largest study to date, reported that the use of HCV viremic grafts in carefully selected aviremic seronegative patients in the DAA era appeared to be efficacious and well tolerated (26). In heart and lung transplantation, DAA treatment prevented the establishment of HCV infection in patients without HCV infection who received a heart or lung transplant from donors with HCV viremia (27).

Despite the increased cost of DAAs, several studies suggest that accepting a HCV seropositive kidney for a aviremic seronegative recipient with DAA therapy is cost-saving (less costly and more effective) compared with remaining on dialysis and waiting for an aviremic seronegative kidney (28,29). A recent study by our group evaluated the financial break-even point. Using current pangenotypic regimens and estimates derived from Medicare payments, results suggest that payers will reduce expenditures if waiting times are reduced by at least 11.5 months (30).

The cost of therapy may be further decreased because shorter courses of therapy may be effective. A recent study reported that using sofosbuvir/velpatasvir for 4 days, with the first dose to be given before transplant, would decrease transmission to 7.5% (31). Although this option would be more appealing to patients and providers, it shifts the cost from insurance companies to the hospital because the medication would be paid under the standard diagnosis-related group payment. This would decrease the hospital’s margins. Developing strategies to address these logistical and financial issues would definitely allow for better use of viremic kidneys.

Donors with HCV infection had favorable transplant factors, including younger age, and lower likelihoods of obesity, diabetes, or hypertension, or death owing to stroke (15). These favorable risk factors should encourage the use of these kidneys by kidney transplant centers and be included in the discussion of organ acceptability with their potential recipients. As suggested by Kling et al., using organs from aviremic seropositive donors at the same rate as aviremic seronegative donors could result in many more organs transplanted per year (32). In particular, studies have demonstrated lower risk of death for accepting a kidney labeled as high risk for disease transmission in comparison with declining and waiting for a normal-risk kidney (33). Nonetheless, as of December 2018 and despite the availability of DAAs, the majority of donor HCV seropositive transplants to aviremic seronegative recipients were performed by a limited group of transplant centers (34).

The KDPI score was higher in donor HCV seropositive kidneys than seronegative kidneys (KDPI score >50%: 87% for HCV seropositive kidneys and 61% for seronegative kidneys). This is related to the fact that HCV status is part of the calculation of the KDPI, with a β coefficient of 0.24 (OR, 1.27), which led to a higher index score despite having better donor characteristics (15,35). With the data on the effectiveness of DAAs and the equivalent allograft survival of these kidneys, a modified version of the KDPI should be developed by omitting or decreasing the β coefficient of HCV in the KDPI formula, to reflect contemporary risk (15,23). This is supported by our finding of higher post-transplant 1-year eGFR for patients receiving kidneys from HCV viremic donors in comparison with those for propensity score-matched patients receiving kidneys from HCV aviremic seronegative donors in the same KDPI group.

This study has several limitations. First, like all retrospective studies, the results of the analyses rely on the accuracy of the recorded data. Second, the NAT results are available in the OPTN data since March 31, 2015. Therefore, we were unable to describe an earlier trend for viremic or aviremic seropositive kidney discards before and at the beginning of the DAA era. Nonetheless, our NAT results contained data from March 31, 2015 to September 30, 2019, allowing us to study the time trend of approximately 5 years in the post-DAA era.

To conclude, discard of kidneys from viremic donors has remarkably decreased in the past 3 years. There is now equivalent use of kidneys from aviremic seropositive donors and kidneys from aviremic seronegative donors. However, kidneys from viremic donors still had 48% higher odds of discard in 2019. There are opportunities for more transplant centers to utilize these organs to expand access to transplant. Broader utilization would benefit from financial coverage for post-transplant DAAs, which would result in wider adoption of protocols to transplant kidneys from viremic donors to uninfected recipients.

Disclosures

T. Alhamad reports consultancy agreements with CareDx, Mallinckrodt, and Veloxis; receiving research funding from Angion, CareDx, Mallinckrodt, and Natera; receiving honoraria from CareDx, Mallinckrodt, Sanofi, and Veloxis; serving as a scientific advisor or member of CareDx and Mallinckrodt; and speakers bureau for CareDx, Sanofi, and Veloxis. S. Anwar reports employment with Seha. D.A. Axelrod reports consultancy agreements with CareDx and Sanofi; ownership interest in CareDx; and receiving honoraria from Sanofi and Veloxis. W.C. Chapman reports receiving research funding from Novartis and having other interests/relationships with Pathfinder. K.L. Lentine reports consultancy agreements with CareDx and speakers bureau for Sanofi. M.A. Schnitzler reports consultancy agreements with CareDx and receiving honoraria from OPTUM. J. Wellen reports consultancy agreements with Veloxis; receiving research funding from Astellas and Novartis; receiving honoraria from Sanofi/Genzyme and Veloxis; and speakers bureau for Sanofi/Genzyme and Veloxis. All remaining authors have nothing to disclose.

Funding

This work was supported by the Mid-America Transplant Clinical Innovation Award grant 032018 (to T. Alhamad, K.L. Lentine, D.A. Axelrod, and M.A. Schnitzler). T. Alhamad is supported by the Foundation of Barnes Jewish Hospital grant 4746. K.L. Lentine is supported by the Jane A. Beckman Endowed Chair in Transplantation.

Supplemental Material

This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.10960720/-/DCSupplemental.

Supplemental Table 1. Donor characteristics for discarded kidneys, by HCV antibody and viremia status.

Supplemental Figure 1. Kidney discards stratified by KDPI category, from 2005 to 2019.

Supplemental Figure 2. Multivariable-adjusted odds ratios for HCV viremic and aviremic seropositive (reference: aviremic seronegative) kidney discards, stratified by KDPI category (<85 and ≥85), from 2005 to 2019.

Supplemental Figure 3. Reasons for viremic kidney discard, by UNOS region (1–11) and year (2015–2019).

Supplemental Figure 4. One-year post-transplant eGFR for kidney transplant patients receiving kidneys from HCV viremic donors versus propensity score-matched aviremic seronegative donors.