The shortage of donor organs for kidney transplant candidates is a leading problem facing the transplant field. Nearly 100 000 patients in the United States are waitlisted for kidney transplantation, yet only 12 000 each year receive deceased donor transplants.1 This scarcity leads to waiting times that exceed 3 to 5 years in many parts of the country. Despite the substantial need for organs, more than 800 kidneys from deceased donors with hepatitis C virus (HCV) infection were discarded in the United States in the year 2016.2 This number does not include those kidneys from HCV-infected potential donors that are never procured. Furthermore, referrals for HCV-infected potential donors are increasing due to the opiate epidemic.3 Elevated discard rates of HCV-infected kidneys were historically driven by low cure rates and major side effects of the legacy HCV treatments, such as interferon, which promotes allograft rejection if administered after transplant.4,5
Given new direct-acting agents (DAA) for HCV infection, our group and others recognized the potential to increase the use of HCV-infected kidneys by offering these kidneys to waitlisted patients without HCV infection.6,7 Several DAA regimens have cure rates greater than 95% in the general population.8-11 A limited but growing body of evidence suggests that DAA treatment can provide high cure rates for immunosuppressed patients with HCV.12-17 However, it is unknown whether HCV-negative kidney transplant candidates would be willing to accept organs from HCV-infected donors. Notably, the pilot Transplanting Hepatitis C Kidneys into Negative Kidney Recipients (THINKER) trial, which involved 10 HCV-negative patients transplanted with HCV-infected kidneys, reported HCV cure among all 10 participants. This trial featured a number of exclusion criteria including patients older than 65 years and prior transplant recipients.17
We aimed to determine how kidney transplant candidates weigh multiple aspects of the kidney-acceptance decision, which may include factors related to their alternative treatment options, kidney quality, and infection risk. Furthermore, because of data demonstrating that individuals who self-identify as black race often have lower trust in physicians than white race individuals, we planned to examine race and trust in physicians as determinants of willingness to accept an HCV-infected kidney.18-20 Given the limited data related to risk of donor-derived HCV infection, we hypothesized that: (1) black patients would have lower willingness to accept an HCV-infected kidney, (2) black race would modify the effect of lower HCV cure rates on willingness to accept an HCV-infected kidney, and (3) less than half of participants overall would be willing to accept an HCV-infected kidney.
MATERIALS AND METHODS
We first enrolled adults undergoing evaluation or reevaluation for kidney transplantation to perform a qualitative study consisting of semi-structured interviews at the University of Pennsylvania (Penn). We then separately enrolled patients in a conjoint study at Penn and Yale University (Yale) transplant centers. Conjoint analysis is an experimental method that presents a series of scenarios and requires participants to make decisions (in this case, about kidney allograft acceptance), which thereby reveal how participants weigh the features in the scenarios. Conjoint analysis shows the relative influence of factors “considered jointly” on participants’ decisions.21
Participants and Administration
Patients were approached while waiting to meet with physicians and social workers during the medical evaluation for kidney transplantation. Inclusion criteria consisted of individuals 18 years or older who could provide informed consent. All individuals lacked evidence of current HCV infection (ie, HCV nucleic acid test negative or, where nucleic acid testing was not performed, most recent HCV antibody negative). We excluded potential participants who were blind or severely visually impaired, and those who could not speak English, because of the attention to specific details of the conjoint scenarios that was required to complete the study. All participants provided written informed consent. The Penn (Protocol 822950) and Yale (Protocol HIC 1605017721) Institutional Review Boards approved the study.
Conjoint Instrument Development
To identify the most relevant variables to patient decision-making, we first performed semistructured interviews with kidney transplant candidates focusing on their experience with end-stage kidney disease and beliefs regarding HCV infection and the idea of accepting an HCV-infected kidney transplant (Items 1 and 2, SDC,http://links.lww.com/TP/B525). We excluded kidney transplant candidates with a history of HCV and/or liver disease. Interviews were conducted at the time of the visit or by phone and were digitally recorded. During 8 clinic days, we invited 33 patients to participate. Twenty enrolled, among whom 7 chose to complete the interview by phone; among these, 3 did not complete interviews despite multiple phone calls. After the 17 interviews were completed, the analysis reached thematic saturation.22,23 The median age was 55 years (interquartile range [IQR], 46-59 years); 55% were black, 40% were white, and 5% were other race. Fifty percent were male.
Two investigators (M.M., A.M.) reviewed all interviews and collaboratively identified the leading concerns related to the decision to accept an HCV-infected kidney. The research staff observed that participants had very limited knowledge about HCV, and some offered incorrect information (Item 3, SDC,http://links.lww.com/TP/B525, provides representative quotes from participants when asked about their knowledge of hepatitis C.) Based on these data, we developed scripted materials for concise education about HCV, potential cure, and the likely impact on waiting time for participants willing to accept an HCV-infected kidney.
Review of the transcripts revealed that the 4 most commonly held concerns regarding HCV-infected kidneys were related to health problems from HCV, kidney quality, expected waiting time for a transplant, and whether the candidate needed to receive chronic dialysis before transplant (Item 4, SDC,http://links.lww.com/TP/B525, provides representative quotes related to these 4 themes). Therefore, we developed 12 scenarios in which we systematically varied: (1) the probability of successfully curing HCV after transplantation (75%, 90%, or 95%), (2) kidney “quality” (a 20-year-old donor or a 60 year-old donor with hypertension), and (3) the anticipated waiting time for an offer of a HCV-negative kidney if the HCV-infected organ was declined (2 or 5 years). We selected 95% as the upper bound of cure rate because of similar HCV cure rates with DAAs in the general population and 75% as a much lower bound because of unknown effects of cure when infection is concurrent with high-dose immunosuppression.8 Donor quality and waiting time for transplant were presented as binary categories because we did not want to overwhelm participants with more than 12 scenarios.
In each of the 12 scenarios, the participant chose to accept the HCV-infected kidney or to decline and continue waiting the specified time until the next HCV negative kidney became available (Item 5, SDC,http://links.lww.com/TP/B525).
The scenarios did not ask the participant to consider a hypothetical dialysis status, the other important decisional factor expressed in pilot interviews. Instead, we concluded that the effect of this participant characteristic was best assessed by potential adjustment in multivariable analysis for whether the participant was actually undergoing chronic dialysis, rather than by asking the participant to imagine being in a dialysis or nondialysis state.
We pilot tested and revised the HCV educational document and conjoint instrument in several iterations among transplant candidates at Penn to assess understanding. Investigators observed that pilot participants expressed difficulty in distinguishing between scenarios, and in some cases, erroneously reported that some scenarios were identical. As a result, in the final trial, scenarios were uniformly ordered, such that each subsequent offer became more advantageous (ie, cure rates went up, waiting time for an uninfected kidney became longer, and donor age got younger), and differences between scenarios were written in bold, underlined, capital text. After these changes, pilot participants were better able to recognize how scenarios varied.
Demographic and Clinical Data
Participants reported education level, current dialysis status, prior transplantation, and if he/she knew anyone with HCV. The questionnaire also included the Trust in Physician Scale, a validated instrument that comprises 11 positive or negative statements about patients’ feelings toward their treating physicians (score range 11 to 55; higher scores indicate greater trust).24,25 We abstracted age, race, sex, liver disease, and evidence of current HCV infection from the electronic medical record.
The sample size was guided by the hypothesized outcome that would require the most participants to detect—the interaction of race-by-HCV cure rate. Using a design in which each respondent completed all 12 scenarios, we selected an effect size of Cohen d = 0.5. Enrolling 128 participants, with a roughly even distribution of black versus nonblack patients, would provide greater than 80% power to detect a race-by-HCV cure rate interaction. We inflated the sample size to more than 170 participants to augment power and account for missing data in a complete case analysis.
All analyses were completed with Stata 14.2 (StataCorp, College Station, TX). We compared distributions of continuous variables (such as age) between scenarios in which respondents accepted or rejected an HCV-infected kidney using the t test or the rank-sum test, as appropriate. A χ2 test was used to compare categorical variables.
For the binary outcome of kidney acceptance, we fit a multivariable regression model with the following covariates: (1) the 3 variables from scenarios, (2) race, (3) trust scale, (4) any variables with associations with the outcome in which P less than 0.15, and (5) any variables whose inclusion in the model changed the magnitude of the odds ratio (OR) associated with scenario variables by >15%. We also tested prespecified interactions between race and cure. Because the distribution of scores for the Trust in Physicians scale was skewed and violated the linearity in the logit assumption required for logistic regression, the scale was transformed into tertiles. The models then included indicator variables for each of the 2 higher tertiles (with the lowest tertile as the reference category). We used the robust variance estimator to account for the fact that the 12 responses of each participant were not independent. The Hosmer-Lemeshow goodness of fit test for calibration was acceptable for all models (P > 0.05). We used the Stata “margins” command with covariates set at their means to estimate acceptance rates for black versus white patients.
Finally, we performed exploratory analyses of the outcome of refusing all offers of HCV-infected kidneys.
Some participants did not complete either all 12 conjoint scenarios or other parts of the data collection instrument, such as the Trust Scale. For analyses of rejecting all organs (Table S1), we limited the sample to individuals who responded to all scenarios (n = 181). For primary analyses, we included participants who completed at least 75% of scenarios (n = 189). We performed sensitivity analyses in which missing data on covariates were assigned extreme values selected from the distribution of values observed in the population without missing data (n = 189). Results were similar to the primary analysis and were not shown.
Figure 1 shows enrollment. From June 2016 to November 2016, among 214 patients who met inclusion criteria and were invited to participate in the conjoint study, 189 consented (n = 63 from Yale and n = 126 from Penn), 181 completed the scenarios, and 8 provided incomplete responses. Table 1 shows participant characteristics. The median age was 51 years (IQR 43, 59) and 62% were male. Self-identified race was 44% white, 45% black, 4% Asian, and 9% other. Only 14% of participants reported knowing someone with HCV.
The Trust in Physician scale was right-skewed, with a median score of 45 (IQR, 41-52; range, 26-55). In exploratory analyses of race and trust, we found only small differences in trust between black and white patients (median, 46 vs 44; P = 0.11). Trust scores were not different between participants at the 2 study sites (median, 45 for Penn vs 48 for Yale; P = 0.27).
For the conjoint scenarios, 29% of respondents reported that they would accept an HCV-infected kidney under all scenarios of varying risk, donor quality, and waiting time, whereas an additional 53% indicated willingness to accept a kidney under at least 1 scenario. Therefore, 82% would accept all or some HCV-infected offers and only 18% refused all offers. Within strata defined by race, 90% of white, 72% of black, and 75% of nonblack, nonwhite participants accepted at least 1 kidney offer.
Overall Conjoint Analysis of Willingness to Accept a Kidney From an HCV-Infected Donor
Contrary to findings from our interviews, in bivariate analyses and multivariate analyses, participant dialysis status was not associated with overall decisions to accept an HCV-infected kidney. Educational attainment, current waitlist status, and knowing someone with HCV were also not associated with willingness to accept an HCV-infected kidney.
In multivariable regression (Table 2), participants were highly influenced by the anticipated cure rate for HCV (OR, 3.49; 95% CI, 2.33-5.24 for 95% cure rate and OR, 2.10; 95% CI, 1.57-2.80 for 90% cure rate vs reference 75% cure rate). Better allograft quality (OR, 2.34; 95% CI, 1.91-2.88 for a 20 year-old vs a 60 year-old donor with hypertension) and longer wait for an HCV-uninfected kidney (OR, 1.43; 95% CI, 1.22-1.67 for 5 years vs 2 years) were also highly associated with accepting an HCV-infected kidney. Among participant attributes, older than 60 years, being in clinic for reevaluation, and prior kidney transplantation were also strongly associated with willingness to accept an HCV-infected kidneys, whereas being a Penn-based participant (vs Yale) was associated with a lower willingness to accept a kidney.
The effect of HCV cure rates were examined for modification by race; we hypothesized that the negative effect of lower cure rates on willingness to accept a kidney would be stronger among black versus white participants. Black race modified the effect of the HCV cure rate (black race × HCV 95% cure rate, P = 0.03). Figure 2 shows acceptance rates at different levels of HCV cure rate. As the HCV cure rate improved, the willingness to accept an HCV-infected kidney rose to a greater degree among white versus black patients. Specifically, as the cure rate rose from 75% to 90% to 95%, the acceptance rate among white participants rose from 53% to 70% to 79%, whereas for black patients, the acceptance rate rose from 49% to 61% to 65%.
We found that trust in physicians was associated with a lower willingness to accept a hepatitis C infected kidney (OR, 0.51; 95% CI, 0.28-0.93 for highest vs lowest trust tertile). Participants at the University of Pennsylvania were also less likely to accept offers of HCV-infected kidneys (OR, 0.52; 95% CI, 0.29-0.94 vs Yale University).
Exploratory Analyses of Refusing All Offers of Kidneys From HCV-Infected Donors
These analyses were limited by the small number of participants with this outcome (n = 33). In bivariate analyses, we found that nonwhite participants (vs white participants) were more than twice as likely to refuse all HCV-infected kidney offers (26% for black vs 25% for nonblack, nonwhite patients, vs 9% for white, P = 0.01). A higher percentage of women versus men refused all HCV-infected kidneys (25% vs 14%), although this did not reach statistical significance (P = 0.06). Age (P = 0.48), trust in physicians (P = 0.54), knowing someone with HCV (P = 0.69), current dialysis (P = 0.18), and being in clinic for initial versus reevaluation (P = 0.24) were not associated with the outcome.
In a logistic regression model adjusted for male sex and study site, black race (OR, 4.31; 95% CI, 1.13-16.3 vs reference white race) was associated with substantially higher odds of refusing all HCV-infected kidneys (Table S1, SDC,http://links.lww.com/TP/B525).
Identifying new options to increase access to kidney transplantation and alleviate the burden of chronic renal disease on the healthcare system is a leading goal of the transplant community.26,27 Pilot studies suggest that wider use of HCV-infected organs among HCV-negative candidates may substantially increase the number of transplants, but only if patients are willing to accept those organs.17 Using conjoint analysis, we found that over 80% of patients would accept HCV-infected kidneys under some or all circumstances, whereas 18% would never accept an HCV-infected kidney under any circumstances. Patients were highly attuned to anticipated cure rates and organ quality when making their decisions.
A key qualitative finding from the pilot and conjoint phases is that most patients acknowledged having little information about HCV, although 14% of the conjoint study population reported knowing someone with HCV infection. To evaluate willingness to consider HCV-infected organs, we first had to develop a standardized approach to educate participants about this virus and treatments for it. Future trials or clinical practice involving HCV-infected organs for solid organ transplantation must account for this lack of knowledge about HCV. We endorse the idea that informed consent processes about HCV and its treatments should first be implemented while patients are on the waitlist and again under the less optimal conditions of an actual organ offer.7 On the other hand, the fact that more than 80% of participants in a multicenter population were willing to accept some offers of HCV-infected organs suggests that the stigma attached to HCV may be overcome through education. Eliciting from transplant candidates their existing concerns or beliefs about donor infections, such as HCV, may be an important element of that educational process.
Consistent with our hypotheses, black race was associated with refusing all kidney offers. In the main analysis, black race also significantly modified the effect of anticipated HCV cure rate on willingness to accept an HCV-infected kidney. Increases in willingness to accept an HCV-infected kidney were smaller among black versus white patients as anticipated HCV cure rates increased to 95%. This finding remained significant despite adjustment for trust in physicians. This finding could be driven by a number of potential explanations, including greater concerns about HCV risks among black patients (even when cure rates are high), or that black patients have better health outcomes than white patients while receiving dialysis.28 We acknowledge that race is a social construct.29 Our finding could have been confounded by other unmeasured attributes such as health literacy. Yet, the fact that racial categories are a social construct does not necessarily diminish the potential influence of race in health outcomes or healthcare decision making, particularly around issues such as transplantation or donor-derived infection specifically that may be viewed by patients as involving serious risk.19,20,30
Notably, we detected an unanticipated association between greater willingness to accept HCV-infected kidneys and lower trust in physicians. On further examination of how study staff presented the trust in physician scale, we note that patients were instructed to respond to items by reflecting upon their attitudes about their “kidney doctors,” rather than doctors generally. We speculate the participants may have been primarily thinking about their referring nephrologists when responding to this scale, because few waitlisted patients would be expected to form relationships as closely with their transplant nephrologists as with their primary medical team. As a result, patients with high trust in their primary medical teams may have been more content with the status quo and less eager to accept the uncertain risks of HCV infection.
Patients older than 60 years and prior transplant recipients had higher odds of being willing to accept HCV-infected organs. Older patients may have been more willing to accept the risk of HCV infection because physicians have counseled them that health deterioration is highly likely on the waiting list or that the long-term risks of HCV posttransplant would not be a major issue given their already advanced age.31 In contrast, prior transplant recipients may have had greater willingness to accept HCV-infected organs because of positive experiences with their initial transplant or familiarity with successful management of donor disease transmission (eg, cytomegalovirus). These findings suggest that investigators designing future studies of using HCV-infected organs among uninfected recipients should consider relaxing inclusion criteria to include these groups of patients. Notably, the THINKER trial and another pilot study that used HCV-infected organs for HCV-negative kidney transplant candidates restricted participation to individuals younger than 65 years.17,32
We also found a significant difference in willingness to accept HCV-infected kidneys between the 2 transplant centers. At both centers, patients were approached after a general educational session about kidney transplantation and types of risk associated with various types of kidney donors. In many cases, patients completed the scenarios after also meeting with physicians and other transplant staff. We speculate that this difference in willingness to accept an HCV-infected kidneys may be due to the variation in the educational process across centers with reference to the benefits of accepting “higher risk” kidney offers or the negative consequences of long waiting times. The centers also had substantial differences in their populations. We consider these differences between center populations to be a strength because of potentially enhanced generalizability versus a single-center study.
There were several limitations. First, results were based on responses to hypothetical scenarios. However, participants visited the transplant center to either join or remain active on the waiting list and were therefore the exact patients for whom the decisions were relevant. Additionally, responses were consistent with findings in THINKER, our pilot trial of transplanting HCV-infected kidneys into HCV-negative patients, where over half of eligible patients who were contacted for participation moved forward with the informed consent process.17 Second, there is a possibility of bias due to the ordering of scenarios, which may have influenced responses. Based on feedback during development of the conjoint instrument, the scenarios were deliberately ordered where each subsequent allograft offer became more attractive for the purpose of highlighting differences between scenarios. Third, unmeasured confounding is possible. We did not, for example, collect information on potentially relevant characteristics including socioeconomic status or health literacy. We also did not restrict the population to those without any evidence of liver disease (other than current HCV infection), although patients with liver disease might plausibly be excluded from trials of HCV-infected kidneys. Additionally, scenarios did not address the PHS increased-risk status of most HCV-infected donors, which would also confer a small probability of HIV transmission.33 We believe that most transplant candidates were aware of the connection between HCV and drug use, because of the standard content of education at both centers. Future studies should explore how PHS increased-risk status of donors might also influence decision-making when donors are known to have HCV. Finally, we only enrolled 20 individuals who were nonblack, nonwhite.
Our study also has considerable strengths. First, we designed the scenarios after substantial pilot testing, during which we incorporated the most important factors that patients articulated about accepting an HCV-infected kidney. Second, the large sample size was sufficient to detect effect modification. Third, conjoint analysis is a statistically powerful approach that is more effective at identifying participant preferences as compared to simple questionnaires.21
In conclusion, over 80% of transplant candidates without HCV infection were willing to accept an HCV-infected kidney in some circumstances. Future trials involving the transplantation of HCV-infected kidneys into uninfected recipients may consider inclusion of prior transplant recipients and individuals older 60 years, given high levels of interest from these groups. Black patients may be particularly sensitive to HCV cure rates when making decisions about accepting an HCV-infected kidney, although a majority of individuals from each racial group was willing to accept at least 1 offer of an HCV-infected kidney.
1. Organ Procurement and Transplantation Network. Data as of July 4, 2017. Available at: https://optn.transplant.hrsa.gov/data/
. Accessed July 4, 2017.
2. Organ Procurement and Transplantation Network (OPTN). Data as of March 3, 2017. Available at: https://optn.transplant.hrsa.gov/data/
. Accessed March 3, 2017.
3. Goldberg DS, Blumberg E, McCauley M, et al. Improving organ utilization to help overcome the tragedies of the opioid epidemic. Am J Transplant
4. Wei F, Liu J, Liu F, et al. Interferon-based anti-viral therapy for hepatitis C virus infection after renal transplantation: an updated meta-analysis. PLoS One
5. Sawinski D, Bloom RD. Novel hepatitis C treatment and the impact on kidney transplantation. Transplantation
6. Reese PP, Abt PL, Blumberg EA, et al. Transplanting hepatitis C-positive kidneys. N Engl J Med
7. Levitsky J, Formica RN, Bloom RD, et al. The American Society of Transplantation Consensus Conference on the use of hepatitis C viremic donors in solid organ transplantation. Am J Transplant
8. Roth D, Nelson DR, Bruchfeld A, et al. Grazoprevir plus elbasvir in treatment-naive and treatment-experienced patients with hepatitis C virus genotype 1 infection and stage 4–5 chronic kidney disease (the C-SURFER study): a combination phase 3 study. Lancet
9. Foster GR, Afdhal N, Roberts SK, et al. Sofosbuvir and velpatasvir for HCV genotype 2 and 3 infection. N Engl J Med
10. Feld JJ, Jacobson IM, Hezode C, et al. Sofosbuvir and velpatasvir for HCV genotype 1, 2, 4, 5, and 6 infection. N Engl J Med
11. Feld JJ, Kowdley KV, Coakley E, et al. Treatment of HCV with ABT-450/r-ombitasvir and dasabuvir with ribavirin. N Engl J Med
12. Sawinski D, Kaur N, Ajeti A, et al. Successful treatment of hepatitis C in renal transplant recipients with direct-acting antiviral agents. Am J Transplant
13. Lin MV, Sise ME, Pavlakis M, et al. Efficacy and safety of direct acting antivirals in kidney transplant recipients with chronic hepatitis C virus infection. PLoS One
14. Sawinski D, Lee DH, Doyle AM, et al. Successful posttransplant treatment of hepatitis C with ledipasvir-sofosbuvir in HIV+ kidney transplant recipients. Transplantation
15. Lubetzky M, Chun S, Joelson A, et al. Safety and efficacy of treatment of hepatitis C in kidney transplant recipients with directly acting antiviral agents. Transplantation
16. Chen K, Lu P, Song R, et al. Direct-acting antiviral agent efficacy and safety in renal transplant recipients with chronic hepatitis C virus infection: a PRISMA-compliant study. Medicine (Baltimore)
17. Goldberg DS, Abt PL, Blumberg EA, et al. Trial of transplantation of HCV-infected kidneys into uninfected recipients. N Engl J Med
18. Wachterman MW, McCarthy EP, Marcantonio ER, et al. Mistrust, misperceptions, and miscommunication: a qualitative study of preferences about kidney transplantation among African Americans. Transplant Proc
19. Armstrong K, McMurphy S, Dean LT, et al. Differences in the patterns of health care system distrust between blacks and whites. J Gen Intern Med
20. Boulware LE, Cooper LA, Ratner LE, et al. Race and trust in the health care system. Public Health Rep
21. Ryan M, McIntosh E, Shackley P. Methodological issues in the application of conjoint analysis in health care. Health Econ
22. Strauss A, Corbin J. Basics of qualitative research: Techniques and procedures for developing grounded theory
ed. Thousand Oaks: Sage Publications; 1997.
23. Malterud K, Siersma VD, Guassora AD. Sample size in qualitative interview studies: guided by information power. Qual Health Res
24. Anderson LA, Dedrick RF. Development of the Trust in Physician scale: a measure to assess interpersonal trust in patient-physician relationships. Psychol Rep
. 1990;67(3 Pt 2):1091–1100.
25. Thom DH, Ribisl KM, Stewart AL, et al. Further validation and reliability testing of the Trust in Physician Scale. The Stanford Trust Study Physicians. Med Care
26. Reese PP, Harhay MN, Abt PL, et al. New solutions to reduce discard of kidneys donated for transplantation. J Am Soc Nephrol
27. Organ Procurement and Organ Procurement and Items 12 and 13. Annual Set of Committee Goals and Progress Report. Available at: https://optn.transplant.hrsa.gov/media/2244/committee_projects.pdf
. Accessed November 12, 2017.
28. Kucirka LM, Grams ME, Lessler J, et al. Association of race and age with survival among patients undergoing dialysis. JAMA
29. Paul-Emile K. The regulation of race in science. George Washington Law Rev
30. Cooper-Patrick L, Gallo JJ, Gonzales JJ, et al. Race, gender, and partnership in the patient-physician relationship. JAMA
31. Schold J, Srinivas TR, Sehgal AR, et al. 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
32. Exploring Renal Transplants Using Hepatitis C Infected Donors for HCV-negative Recipients (EXPANDER-1). Available at: https://clinicaltrials.gov/ct2/show/NCT02781649
. Accessed November 12, 2017.
33. Kucirka LM, Sarathy H, Govindan P, et al. Risk of window period HIV infection in high infectious risk donors: systematic review and meta-analysis. Am J Transplant