Abstract
Background: Pediatric (age < 18 years) kidney transplant (KT) candidates face increasingly complex choices. The 2014 kidney allocation system nearly doubled wait times for pediatric recipients. Given longer wait times and new ways to optimize compatibility, more pediatric candidates may consider kidney-paired donation (KPD). Motivated by this shift and the potential impact of innovations in KPD practice, we studied pediatric KPD procedures in the US from 2008 to 2021. Methods: We describe the characteristics and outcomes of pediatric KPD recipients with comparison to pediatric non-KPD living donor kidney transplants (LDKT), pediatric LDKT recipients, and pediatric deceased donor (DDKT) recipients. Results: Our study cohort includes 4987 pediatric DDKTs, 3447 pediatric non-KPD LDKTs, and 258 pediatric KPD transplants. Fewer centers conducted at least one pediatric KPD procedure compared to those that conducted at least one pediatric LDKT or DDKT procedure (67, 136, and 155 centers, respectively). Five centers performed 31% of the pediatric KPD transplants. After adjustment, there were no differences in graft failure or mortality comparing KPD recipients to non-KPD LDKT, LDKT, or DDKT recipients. Discussion: We did not observe differences in transplant outcomes comparing pediatric KPD recipients to controls. Considering these results, KPD may be underutilized for pediatric recipients. Pediatric KT centers should consider including KPD in KT candidate education. Further research will be necessary to develop tools that could aid clinicians and families considering the time horizon for future KT procedures, candidate disease and histocompatibility characteristics, and other factors including logistics and donor protections.
| Original language | English (US) |
|---|---|
| Article number | e14657 |
| Journal | Pediatric Transplantation |
| Volume | 28 |
| Issue number | 1 |
| DOIs | |
| State | Published - Feb 2024 |
Keywords
- donor exchange
- living donation
- paired donation
- pediatric
ASJC Scopus subject areas
- Pediatrics, Perinatology, and Child Health
- Transplantation
Fingerprint
Dive into the research topics of 'Unrecognized opportunities: The landscape of pediatric kidney-paired donation in the United States'. Together they form a unique fingerprint.Cite this
- APA
- Standard
- Harvard
- Vancouver
- Author
- BIBTEX
- RIS
In: Pediatric Transplantation, Vol. 28, No. 1, e14657, 02.2024.
Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - Unrecognized opportunities
T2 - The landscape of pediatric kidney-paired donation in the United States
AU - Verbesey, Jennifer
AU - Thomas, Alvin G.
AU - Waterman, Amy D.
AU - Karhadkar, Sunil
AU - Cassell, Victoria R.
AU - Segev, Dorry L.
AU - Hogan, Julien
AU - Cooper, Matt
N1 - Funding Information: Dr. Segev reports consulting and speaking honoraria from Sanofi, Novartis, CLS Behring, Jazz Pharmaceuticals, Veloxis, Mallinckrodt, and ThermoFisher Scientific. New York University Langone Health receives institutional research grants from the National Kidney Registry that supports the Center for Surgical & Transplant Applied Research (C‐STAR) directed by Dr. Segev. No other disclosures are reported. Funding Information: This study was supported, in part, by the following grants from the National Institutes of Health (NIH): T32HL007055 (for Alvin Thomas), F99AG073565 (PI: Alvin Thomas), and K24AI144954 (PI: Dorry Segev). Additionally, Drs. Segev and Waterman are supported, in part, by institutional grant funding award to C-STAR from the NKR, and Dr. Waterman receives funding support from Health Resources and Services Administration (HRSA) Grant R39OT45484. Role of the sponsors: The NIH and HRSA had no direct role in this study. The NKR was involved in the data collection and review of the manuscript. The NKR was not involved in the analysis or interpretation of the data nor the decision to submit the manuscript for publication. National transplantation data, including pediatric DDKT, LDKT, and KPD rates, are routinely collected from transplant centers and reported to the United Network for Organ Sharing (UNOS), which acts as the Organ Procurement and Transplantation Network (OPTN) contractor for the US Department of Health and Human Services. These data are systemized by the SRTR as a contractor for the Health Resources and Services Administration. As part of their mandate, SRTR supplements national transplantation data with external sources, such as death data from the National Death Index.14 The NKR is a non-profit, 501(c) organization that facilitates KPD in the US across ~100 partner transplant centers. The characteristics and unique features of the NKR dataset of all KPD transplants have been discussed previously.15–19 Since its inception in 2007, the NKR has facilitated over 7300 living donor kidney transplants. As such, the NKR is the largest network for KPD in the US and the world.20 Under a data use agreement between New York University (NYU) Langone Heath's Center for Surgical and Transplant Applied Research (C-STAR), the SRTR and NKR datasets were linked using unique, encrypted, person-level identifiers. The quality of the linkage was assessed by using redundantly collected data fields such as transplant date, recipient and donor blood type, and recipient and donor sex. By linking SRTR and NKR data, KPD misclassification risk is reduced. We restricted our study population to transplants that occurred between February 2008 and December 2021. All recipients were followed until June 2022 (i.e., a minimum of 6 months of post-transplant follow-up) using the December 2022 SRTR standard analysis file. To align with KAS' definition of pediatric recipients,21 inclusion criteria for the study defined pediatric kidney recipients as those who were < 18 years from birth on the day of the transplant. We excluded multi-organ transplants. This study has been approved by the NYU Grossman School of Medicine Institutional Review Board (i22-01407). The clinical and research activities of this study are consistent with the Declaration of Helsinki and the Declaration of Istanbul. National transplantation data, including pediatric DDKT, LDKT, and KPD rates, are routinely collected from transplant centers and reported to the United Network for Organ Sharing (UNOS), which acts as the Organ Procurement and Transplantation Network (OPTN) contractor for the US Department of Health and Human Services. These data are systemized by the SRTR as a contractor for the Health Resources and Services Administration. As part of their mandate, SRTR supplements national transplantation data with external sources, such as death data from the National Death Index.14 The NKR is a non-profit, 501(c) organization that facilitates KPD in the US across ~100 partner transplant centers. The characteristics and unique features of the NKR dataset of all KPD transplants have been discussed previously.15–19 Since its inception in 2007, the NKR has facilitated over 7300 living donor kidney transplants. As such, the NKR is the largest network for KPD in the US and the world.20 Under a data use agreement between New York University (NYU) Langone Heath's Center for Surgical and Transplant Applied Research (C-STAR), the SRTR and NKR datasets were linked using unique, encrypted, person-level identifiers. The quality of the linkage was assessed by using redundantly collected data fields such as transplant date, recipient and donor blood type, and recipient and donor sex. By linking SRTR and NKR data, KPD misclassification risk is reduced. We restricted our study population to transplants that occurred between February 2008 and December 2021. All recipients were followed until June 2022 (i.e., a minimum of 6 months of post-transplant follow-up) using the December 2022 SRTR standard analysis file. To align with KAS' definition of pediatric recipients,21 inclusion criteria for the study defined pediatric kidney recipients as those who were < 18 years from birth on the day of the transplant. We excluded multi-organ transplants. This study has been approved by the NYU Grossman School of Medicine Institutional Review Board (i22-01407). The clinical and research activities of this study are consistent with the Declaration of Helsinki and the Declaration of Istanbul. The primary exposure of interest was the mode of pediatric kidney transplant (KPD, non-KPD LDKT, and DDKT). Pediatric kidney transplants were identified to be KPDs when they had either an SRTR-reported donor relationship related to KPD (codes 9, 11, and 12) or linkage from NKR (confirmed KPD). We did not include SRTR-reported non-directed donors (code 10) unless they were linked by NKR since they could have donated directly to the waitlist.8,17 We note that not counting non-directed donors toward KPD likely results in a quantitative bias; the number of KPDs we report may be an underestimate since some portion of non-directed donors likely participated in KPD. This bias should be reduced over time as the proportion of KPD facilitated by NKR increases (since NKR-linked non-directed donations are confirmed KPD transplants). We examined transplant outcomes including death-censored graft failure (DCGF) and mortality. As in other studies, DCGF was defined as the earliest of three events: resumption of maintenance dialysis, relisting for kidney transplant, or retransplantation. Transplant centers' reports of DCGF to the OPTN are supplemented by Centers for Medicare and Medicaid Services Form 2728. Likewise, transplant center reports of death are supplemented by the Department of Health Statistics Death Master File. We describe differences between KPD and non-KPD LDKTs but refrain from quantifying differences using statistical tests to avoid errors in inference from multiple testing or information bias (e.g., confounding).22 We report categorical variables as percentages and continuous variables as the median value followed by the limits of the interquartile range (IQR, 25% and 75% percentile values). Using a time-to-event framework, we assess DCGF and mortality outcomes by transplant mode. We used the Kaplan-Meier estimator to plot survival functions. We compare survival functions using the log-rank test with pre-specified α of.05 to indicate a statistically significant difference. Additionally, we used Cox proportional hazard models to estimate hazard ratios (HR) comparing KPD recipients to controls. We present results from three models that represent alternative assumptions regarding the underlying relationship between transplant characteristics. Model 1 accounts for recipient factors only: sex, age, Black race, public insurance status, years of dialysis, height, weight, and preemptive status. Model 2 accounts for recipient factors as well as donor factors: sex, age, Black race, BMI ≥ 30 kg/m2 (body mass index), and LKDPI (living kidney donor profile index). The final model accounts for the same factors as Model 2 with the addition of transplant factors: ABO incompatible transplant and zero HLA mismatches for A, B, and DR loci. We do not include ABO incompatible transplants when comparing KPD to deceased donor transplant recipients (violation of exchangeability). Race, as reported by transplant centers, may reflect self-identified or interviewer-identified racial identity.23 The data collection forms do not provide further insight into this characteristic. We hypothesize that race is a potential confounder based on socialized features of race in the US context, which are known to impact health.24–26 For pediatric recipients, we use height and weight instead of BMI, since body measurements may perform better in pediatric patient populations with diseases known to impact height and/or weight.27 Reports of sex were limited to male or female without further insight into how centers assessed this characteristic. Living kidney donor profile index is an extension of the kidney donor profile index (KDPI) developed by Massie and colleagues to quantify living donor organ quality as deceased donor organ quality.28 KDPI, for deceased donors, ranges from 0 to 100 (lower is a higher quality organ). LKDPI theoretically ranges from −100 to 100, with the 0-100 range directly matching the equivalent value on the KDPI scale. Negative LKDPI values denote an organ of higher quality than any deceased donor organ.18 Given low levels of missingness among covariates, we assumed that missingness followed a missing at random process. Considering this, we performed multiple imputations by chained equations to generate 10 imputed datasets.29 Missing binary variables were imputed using augmented logistic regression. Continuous variables were imputed using predictive mean matching among 10 nearest neighbor values.30 To account for the characteristics listed above, we use stabilized inverse probability weights after the imputation process.31 We winsorized weights below the first and above 99th percentile of weights.32,33 To account for potential center-level differences in the baseline hazard, we performed Cox regression with strata for each center. We report the adjusted hazard ratios, 95% confidence intervals, and p-values. All analyses were performed using Stata 17/MP for Linux. The data that support the findings of this study are available from the National Kidney Registry. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of the National Kidney Registry. Dr. Segev reports consulting and speaking honoraria from Sanofi, Novartis, CLS Behring, Jazz Pharmaceuticals, Veloxis, Mallinckrodt, and ThermoFisher Scientific. New York University Langone Health receives institutional research grants from the National Kidney Registry that supports the Center for Surgical & Transplant Applied Research (C-STAR) directed by Dr. Segev. No other disclosures are reported. Between February 2008 and December 2021, 8692 pediatric transplants met our study's inclusion and exclusion criteria (Table 1). Of these, there were 4987 pediatric DDKTs performed at 155 transplant centers and 3705 LDKTs performed at 136 centers. Among LDKTs, only 258 (7%) were pediatric KPD transplants. From 2008 to 2021, the total number of pediatric KPD transplants increased nearly every year (post-hoc linear-by-linear trend test p <.01). The proportion of pediatric KPD transplants facilitated by NKR compared to all pediatric KPDs went from a low of 0% in 2008 to 88% in 2021 (Figure 1). Overall, 120 (47%) of all pediatric KPDs during this period were facilitated by the NKR. Abbreviations: BMI, body mass index; DDKT, deceased donor kidney transplant; KPD, kidney-paired donation; LDKT, living donor kidney transplant. Compared to pediatric non-KPD LKDT recipients, pediatric KPD recipients were more likely to be Black (15% vs. 8%), have received a previous transplant (16% vs. 7%), and have a PRA >80 (12% vs. 3%). Using the LKDPI calculator to denote donor organ quality, pediatric KPD and pediatric non-KPD living donor organs were similar in quality. However, both transplant modes had lower observed LKPI/KDPI compared to pediatric DDKT (2% vs. 13%). KPD and non-KPD LDKT recipients were observed to have higher rates of preemptive transplant compared with DDKT recipients (37.6% and 40.8% vs. 24.4%) (Table 1). A total of 67 transplant centers performed at least one pediatric KPD transplant since 2008. Among these centers, five centers performed at least 10 KPD transplants and account for 31% of all the pediatric KPD transplants that occurred. We did not observe any differences in overall pediatric transplant volume between centers that used KPD at least once and those that did not. Pediatric KPD transplants occurred at both small and large pediatric transplant centers across the country. Between February 2008 and December 2021, 8692 pediatric transplants met our study's inclusion and exclusion criteria (Table 1). Of these, there were 4987 pediatric DDKTs performed at 155 transplant centers and 3705 LDKTs performed at 136 centers. Among LDKTs, only 258 (7%) were pediatric KPD transplants. From 2008 to 2021, the total number of pediatric KPD transplants increased nearly every year (post-hoc linear-by-linear trend test p <.01). The proportion of pediatric KPD transplants facilitated by NKR compared to all pediatric KPDs went from a low of 0% in 2008 to 88% in 2021 (Figure 1). Overall, 120 (47%) of all pediatric KPDs during this period were facilitated by the NKR. Abbreviations: BMI, body mass index; DDKT, deceased donor kidney transplant; KPD, kidney-paired donation; LDKT, living donor kidney transplant. Compared to pediatric non-KPD LKDT recipients, pediatric KPD recipients were more likely to be Black (15% vs. 8%), have received a previous transplant (16% vs. 7%), and have a PRA >80 (12% vs. 3%). Using the LKDPI calculator to denote donor organ quality, pediatric KPD and pediatric non-KPD living donor organs were similar in quality. However, both transplant modes had lower observed LKPI/KDPI compared to pediatric DDKT (2% vs. 13%). KPD and non-KPD LDKT recipients were observed to have higher rates of preemptive transplant compared with DDKT recipients (37.6% and 40.8% vs. 24.4%) (Table 1). A total of 67 transplant centers performed at least one pediatric KPD transplant since 2008. Among these centers, five centers performed at least 10 KPD transplants and account for 31% of all the pediatric KPD transplants that occurred. We did not observe any differences in overall pediatric transplant volume between centers that used KPD at least once and those that did not. Pediatric KPD transplants occurred at both small and large pediatric transplant centers across the country. We observed higher rates of delayed graft function when comparing pediatric KPD to non-KPD LDKT recipients (4.3% vs. 3.1%; Table 1), but this difference was not statistically significant (post-hoc Wilcoxon ran-sum p-value <.001). We observed higher rates of delayed graft function (7.9%) and cold ischemic times (11.9 vs. 2.3 h) comparing DDKT transplants to KPD and non-KPD LDKT transplants. The absolute number of patients without immediate function was low in this study population. We did not observe statistical differences in the mortality risk curve comparing KPD to non-KPD LDKT (Figure 2). However, we did observe a statistically significant difference in graft failure comparing KPD to non-KPD LDKT pediatric recipients. In the unadjusted Kaplan-Meier analysis, KPD graft failure risk was higher than non-KPD graft failure risk. After adjustment, this association was not statistically significant. After adjusting for recipients' characteristics, there was no significant difference in the risk of graft loss between the two groups (HR 1.23 [0.86, 1.78]; Table 2). After adjustment, there were no statistically significant differences in graft failure or mortality between KPD and non-KPD LDKT, but HR values were above 1 (Table 2). When we compared KPD to DDKT, there were no observed differences in graft failure or mortality risk (Figure 3). After adjustment for at least recipient and donor factors, HR values were below 1 (Table 3). Note: All models adjusted for recipient factors: sex, age, Black race, public insurance status, years of dialysis, height, weight, and preemptive status. Donor factors were donor sex, age, Black race, BMI ≥ 30 kg/m2, and LKDPI. Transplant factors were ABO incompatible transplant and zero HLA mismatches for A, B, and DC loci. Note: All models adjusted for recipient factors: sex, age, Black race, public insurance status, years of dialysis, height, weight, and preemptive status. Donor factors were donor sex, age, Black race, BMI ≥ 30 kg/m2, and LKDPI. Transplant factors were zero HLA mismatches for A, B, and DC loci. Pediatric KT candidates and their families face increasingly complex choices under the kidney allocation system (KAS).1–3 KAS policies aimed to improve longevity matching between donors and recipients and increase access for underserved populations.4 While KAS successfully reduced deceased donor kidney transplant (DDKT) disparities between Black and White adult recipients (and some pediatric recipients), the overall rate of pediatric KT decreased.3,5,6 Post-KAS implementation, pediatric KT candidates experienced longer DDKT wait times (median 7-11 months) compared to pre-KAS wait times (median 3-4 months).7,8 Based on organ offer analyses, longer post-KAS wait times are partly explained by higher organ offer rejection rates. This suggests that pediatric transplant centers are less satisfied with the quality of post-KAS organ offers compared to pre-KAS offers.2 Given these changes, living donor kidney transplantation (LDKT), especially kidney-paired donation (KPD), has become an increasingly important mode of KT.9 Pediatric patients and families may pursue LDKT for potentially shorter wait times, longer graft survival, and better immunologic matching. The practice of LDKT has shifted over the past decade with KPD now representing a significant portion of the annual LDKTs in the US.7,9 Innovations in kidney-paired donation (KPD) also provide new opportunities for pediatric candidates and their potential donors. Examples include advanced donation, where living donors can donate earlier than the recipient, and voucher donation, where someone can donate and appoint others, including immediate family members, to receive a living donor transplant in the future.10–12 Advance donation includes initial donation and a “kidney voucher” for the recipient to be prioritized for a living donor organ used when ready. This allows for a “decoupling” of a donor-recipient pair so that the donor can proceed at an optimal time, and is particularly helpful with pediatric patients so that the parent and child are not recovering from surgery simultaneously. Unfortunately, most KPD-published research focuses on adult recipients' characteristics and outcomes. This gap may impact the guidance available for pediatric KT candidates about whether to pursue DDKT, LDKT, or KPD, especially in the context of recent KPD innovations. For example, the National Kidney Registry (NKR) now offers a Donor Shield program that aims to remove disincentives to living donation by providing lost wage reimbursement for donors, travel and lodging expenses, prioritization for a future kidney offer through the NKR if the donor ever suffers kidney failure, and reimbursement for complications associated with donor nephrectomy.13 This could help the family of a pediatric KT candidate considering a parent donor. The parent's nephrectomy could occur earlier with sufficient time for recovery prior to the child's transplant procedure. The ability to decouple these surgeries in time would reduce caregiver burden for this family. However, weighing this option against counterfactual KPD, LDKT, or DDKT options may be impossible given limited data on pediatric KPD experiences and outcomes. Pediatric and adult candidates weigh different sets of conditions and time horizons. While recent KPD innovations will likely change the future uptake and experience of pediatric KPD, clinicians and pediatric KT candidates require pediatric-specific evidence to make informed transplant decisions. Thus, we performed a retrospective study of KPD, LDKT, and DDKT transplants performed in the US between 2008 and 2021 using data from the Scientific Registry of Transplant Recipients (SRTR) and the NKR. We describe the frequency of transplants and patient characteristics. We also compare rates of death-censored graft failure (DCGF) and mortality of pediatric KPD recipients to pediatric non-KPD LDKTs and DDKTs. Finally, we examined variations in KPD participation among transplant centers. Funding Information: This study was supported, in part, by the following grants from the National Institutes of Health (NIH): T32HL007055 (for Alvin Thomas), F99AG073565 (PI: Alvin Thomas), and K24AI144954 (PI: Dorry Segev). Additionally, Drs. Segev and Waterman are supported, in part, by institutional grant funding award to C‐STAR from the NKR, and Dr. Waterman receives funding support from Health Resources and Services Administration (HRSA) Grant R39OT45484. Role of the sponsors: The NIH and HRSA had no direct role in this study. The NKR was involved in the data collection and review of the manuscript. The NKR was not involved in the analysis or interpretation of the data nor the decision to submit the manuscript for publication. Publisher Copyright: © 2023 The Authors. Pediatric Transplantation published by Wiley Periodicals LLC.
PY - 2024/2
Y1 - 2024/2
N2 - Background: Pediatric (age < 18 years) kidney transplant (KT) candidates face increasingly complex choices. The 2014 kidney allocation system nearly doubled wait times for pediatric recipients. Given longer wait times and new ways to optimize compatibility, more pediatric candidates may consider kidney-paired donation (KPD). Motivated by this shift and the potential impact of innovations in KPD practice, we studied pediatric KPD procedures in the US from 2008 to 2021. Methods: We describe the characteristics and outcomes of pediatric KPD recipients with comparison to pediatric non-KPD living donor kidney transplants (LDKT), pediatric LDKT recipients, and pediatric deceased donor (DDKT) recipients. Results: Our study cohort includes 4987 pediatric DDKTs, 3447 pediatric non-KPD LDKTs, and 258 pediatric KPD transplants. Fewer centers conducted at least one pediatric KPD procedure compared to those that conducted at least one pediatric LDKT or DDKT procedure (67, 136, and 155 centers, respectively). Five centers performed 31% of the pediatric KPD transplants. After adjustment, there were no differences in graft failure or mortality comparing KPD recipients to non-KPD LDKT, LDKT, or DDKT recipients. Discussion: We did not observe differences in transplant outcomes comparing pediatric KPD recipients to controls. Considering these results, KPD may be underutilized for pediatric recipients. Pediatric KT centers should consider including KPD in KT candidate education. Further research will be necessary to develop tools that could aid clinicians and families considering the time horizon for future KT procedures, candidate disease and histocompatibility characteristics, and other factors including logistics and donor protections.
AB - Background: Pediatric (age < 18 years) kidney transplant (KT) candidates face increasingly complex choices. The 2014 kidney allocation system nearly doubled wait times for pediatric recipients. Given longer wait times and new ways to optimize compatibility, more pediatric candidates may consider kidney-paired donation (KPD). Motivated by this shift and the potential impact of innovations in KPD practice, we studied pediatric KPD procedures in the US from 2008 to 2021. Methods: We describe the characteristics and outcomes of pediatric KPD recipients with comparison to pediatric non-KPD living donor kidney transplants (LDKT), pediatric LDKT recipients, and pediatric deceased donor (DDKT) recipients. Results: Our study cohort includes 4987 pediatric DDKTs, 3447 pediatric non-KPD LDKTs, and 258 pediatric KPD transplants. Fewer centers conducted at least one pediatric KPD procedure compared to those that conducted at least one pediatric LDKT or DDKT procedure (67, 136, and 155 centers, respectively). Five centers performed 31% of the pediatric KPD transplants. After adjustment, there were no differences in graft failure or mortality comparing KPD recipients to non-KPD LDKT, LDKT, or DDKT recipients. Discussion: We did not observe differences in transplant outcomes comparing pediatric KPD recipients to controls. Considering these results, KPD may be underutilized for pediatric recipients. Pediatric KT centers should consider including KPD in KT candidate education. Further research will be necessary to develop tools that could aid clinicians and families considering the time horizon for future KT procedures, candidate disease and histocompatibility characteristics, and other factors including logistics and donor protections.
KW - donor exchange
KW - living donation
KW - paired donation
KW - pediatric
UR - http://www.scopus.com/inward/record.url?scp=85182692426&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85182692426&partnerID=8YFLogxK
U2 - 10.1111/petr.14657
DO - 10.1111/petr.14657
M3 - Article
C2 - 38317337
AN - SCOPUS:85182692426
SN - 1397-3142
VL - 28
JO - Pediatric Transplantation
JF - Pediatric Transplantation
IS - 1
M1 - e14657
ER -