Kidney Transplant Outcomes Associated With the Use of Increased Risk Donors in Children

Sarah J. Kizilbash; Michelle N. Rheault; Qi Wang; David M. Vock; Srinath Chinnakotla; Tim Pruett; Blanche M. Chavers


American Journal of Transplantation. 2019;19(6):1684-1692. 

In This Article


This is the first and the largest pediatric study to examine IRD kidney transplant rates in children and to report the outcomes associated with IRD transplants. Between 2005 and 2015, 6% of all deceased pediatric kidney transplants utilized IRD organs. The number of pediatric IRD transplants increased from 3.4-6.9% to 10.7-13.2% after the implementation of the new CDC guidelines. Our study did not find any difference in the overall patient or graft survival between IRD and non-IRD pediatric kidney transplant recipients. However, IRD recipients had superior patient survival compared with remaining on the waitlist and declining subsequent IRD kidneys (P = .018). We found no difference in the number of deaths or graft losses due to infections. Furthermore, IRD recipients were more likely to have blood group "O" and higher cPRA compared with non-IRD recipients.

IRDs are at risk of inadvertently transmitting HIV, hepatitis B or C infections to recipients through transplant due to the inherent limitations of ELISA and NAT in detecting window period infections. The duration of the window period varies with the test. Compared with ELISA, NAT detects HIV virus 12 to 13 days earlier, hepatitis B virus 21 to 36 days earlier, and hepatitis C virus 26 to 60 days earlier.[18] Following the CDC's updated guidelines, OPTN board of directors mandated NAT testing for hepatitis C for all donors and NAT or antigen/antibody combination testing for HIV for IRDs.3 The risk of an undetected infection is negligible for exposures that occur more than one month prior to donation. A systemic review and meta-analysis by Kucirka et al. showed that the risk of undetected hepatitis C window period infection ranged from 0.027 to 32.4 per 10 000 IRDs,[5] and the risk of undetected HIV infection ranged from 0.04 to 4.9 per 10 000 IRDs.[6] Given the small magnitude of risk, it would be reasonable to argue that in the setting of a severe organ shortage indiscriminate rejection of these organs should be reconsidered.

Under the new CDC guidelines, IRDs constitute one-fifth of the deceased donor pool in the United States; however, IRD transplant comprised only 13% of the deceased donor pediatric kidney transplants in our study for the year 2015, indicating underutilization. This finding is consistent with adult data. Duan et al, in their study of 45 112 deceased donors, found that IRDs were 8.2% less likely to be used for transplant compared with non-IRDs.8 Similarly, Volk et al, through their study of 35 770 deceased donor kidneys using the OPTN data, demonstrated an absolute difference of 7% in the utilization of IRD versus non-IRD kidneys, and indicated that IRD designation resulted in 313 fewer kidney transplants per year in the United States.[19] There are several reasons for IRD underutilization including the fear of transmitting a stigmatized infection, providers' fear of legal repercussions and hospital pressures.[20] In 2007, an IRD, who had tested negative on ELISA but retrospectively tested positive on NAT, transmitted HIV and hepatitis C to four solid organ recipients. A survey of 422 adult transplant surgeons revealed that 31.6% changed their practice following this event, and 41.7% of those who changed practice decreased the utilization of IRDs.[20] Lack of awareness regarding IRDs and misconceptions among providers and patients also results in low utilization rates.[21,22] A survey of 519 nephrologists revealed that 35.4% were unfamiliar with IRDs and 39% felt ill-prepared to counsel their patients regarding IRDs.[23] Ros et al, through their study of focus groups, illustrated that patients' willingness to consider IRDs could be increased following a brief education session on donor screening.[24] Hence, studies like ours are needed to inform patients and providers about the survival benefit of an IRD transplant compared to remaining on the waitlist for a non-IRD deceased donor kidney, and to promote IRD utilization (particularly for candidates with long expected wait times).

We found no difference in overall graft survival, death-censored graft survival, or patient survival between IRD and non-IRD recipients. Bowring et al demonstrated that IRD transplants offered a survival benefit for kidney recipients compared with remaining on the waitlist.[10] Our results also indicate superior patient survival in IRD recipients compared with remaining on the waitlist and declining future IRD kidney offers (but possibly accepting non-IRD deceased donor kidney offers) (P = .018). Like us, Sahulee et al did not find any difference in patient survival or early rejection between IRD versus non-IRD pediatric heart transplant recipients.[25] Similarly, Gaffey et al found no difference in acute rejection and patient survival between IRD and non-IRD adult heart transplant recipients.[11] Comparable outcomes between IRD and non-IRD recipients and superior patient survival outcomes in IRD recipients compared with remaining on the waitlist argue in favor of IRD transplants in children, particularly for children whose characteristics portend long waitlist times, such as high cPRA or blood group "O."

Chow et al showed that transplant candidates with high PRA and an expected wait time of >1 to 3 years for a non-IRD offer were most likely to benefit from an IRD transplant.[26] Although pediatric recipients are expected to have shorter wait times due to pediatric allocation priority, nearly 20% of pediatric waitlisted candidates are highly sensitized and, approximately 20% wait longer than 3 years for a kidney.[1] Hence, a sizeable proportion of pediatric waitlisted candidates fits the phenotype identified by Chow and could benefit from accepting an IRD kidney. A timely transplant and minimal waitlist time are critical for children to optimize their growth and development.[27]

The risk of virus transmission through transplant is low. A retrospective study of 11 188 pediatric transplant recipients, who were transplanted from 2008 to 2013, found zero cases of HIV and hepatitis C transmission to any solid organ recipient and one case of hepatitis B transmission to a liver recipient through transplant.[28] We queried UNOS and found no documentation of donor derived HIV/hepatitis B/hepatitis C infections in pediatric kidney transplant recipients (queried May 22, 2018). Although virus transmission data were not available in SRTR, we found no evidence of posttransplant hepatitis B treatment in any of the IRD recipients. Additionally, we found no difference in graft loss or death from infections between IRD and non-IRD recipients. Adult studies also found no evidence of HIV/hepatitis/hepatitis C transmission to recipients through IRD organ transplants.[29,30] While the risk of virus transmission is not zero, our national experience shows that risk is miniscule.

Compared with non-IRDs, IRDs are more likely to be younger, male, and black, have lower KDPI scores, have anoxia as their cause of death and are less likely to be expanded criteria donor.[4,9,29] We did not find any differences in age, KDPI scores or expanded criteria characteristics. This is likely because pediatric recipients are preferentially offered higher quality kidneys with lower KDPI scores (under share 35 and new kidney allocation systems). Sahulee et al in their retrospective study of 2782 pediatric heart transplant recipients found that IRDs had a significantly higher BMI;[25] however, we did not find any difference in the BMI between the two groups.

Bowring et al studied 104 998 adult candidates who were offered IRD organs and found that IRD recipients were more likely to be white, highly sensitized (cPRA ≥ 80%), have glomerulonephritis as the cause of ESRD, and have longer wait time on dialysis.[10] In contrast, Gaffey et al found no significant difference in pretransplant dialysis time.[11] We found no difference in race or mean pretransplant dialysis duration between IRD and non-IRD recipients. However, IRD recipients were more likely to have higher cPRA and blood group "O," characteristics that are associated with longer wait times.[31,32]

Our study has several limitations. Firstly, our data lacked granularity and we were unable to measure the rates of virus transmission through transplant from IRDs to kidney recipients. However, our inability to measure transmission rates does not compromise the validity of our survival analyses. Secondly, some kidneys labeled non-IRD would meet the current definition of IRD due to the change in definition over time. Thirdly, our inferences were limited by a short duration of follow-up; only 3 years of follow-up data were available for patients who were transplanted in 2015. Research has shown that survival differences between hepatitis C infected and uninfected kidney transplant recipients do not manifest until after 10 years of follow-up.[33] Similarly, a 5-year graft survival difference may not be seen between HIV infected (monoinfection) and uninfected kidney transplant recipients.[34] Having said this, a longer follow up in our study would have had limited generalizability to the present day due to the change in IRD definition over time.

In conclusion, IRDs are underutilized in pediatric kidney transplant recipients despite constituting a significant proportion of the deceased donor pool. IRD kidney transplants are associated with superior patient survival compared with remaining on the waitlist for a non-IRD deceased donor kidney. We found no difference in patient or graft survival between IRD and non-IRD recipients, which indicates that IRD organs may safely be used in pediatric transplant recipients. NAT testing has significantly reduced the length of window period for all three viruses; hence, the risk of a window period infection is particularly small if the high-risk activity occurs more than 3 weeks prior to donation. Moreover, vaccination against hepatitis B and curative direct acting antiviral therapy for hepatitis C have substantially increased the safety of IRD organs against these infections.[33] In light of our findings and these facts, we recommend that IRD organs be considered for transplant in pediatric candidates, particularly for those who are expected to wait a long time for a non-IRD kidney. Additional studies are needed to examine the incidence of donor-derived HIV/hepatitis B/hepatitis C infections in pediatric IRD solid organ transplant recipients and to study the impact of IRD organ transplant on quality of life.